ISOLATION OF y-AMINOBUTYRATE (GABA) UTILIZING MUTANT STRAINS AND CHARACTERIZATION OF FACTORS SUFFICIENT FOR GABA UTILIZATION AS A CARBON AND NITROGEN SOURCE. ISOLATION OF y-AMINOBUTYRATE (GADA) UTILIZING MUTANT STRAINS AND CHARACTERIZATION OF FACTORS SUFFICIENT FOR GADA UTILIZATION AS A CARBON AND NITROGEN SOURCE

By

SAROSH YOUSUF, B.Sc. (Hons.)

A Thesis

Submitted to the School of Graduate Studies

in Partial Fulfilment of the Requirements

for the Degree

Master of Science

McMaster University

© Copyright by Sarosh Yousuf, January 2001 MASTER OF SCIENCE McMaster University (Biology) Hamilton, Ontario

TITLE: Isolation ofy-aminobutyrate (GABA) utilizing mutant strains and characterization of factors sufficient for GABA utilization as a carbon and nitrogen source.

AUTHOR: Sarosh Yousut: RSc. (Hons) (Shippensburg University)

SUPERVISOR: Dr. H.E. Schellhorn

NUMBER OF PAGES: xi, 94

ii Abstract

We have previously found that the gab operon ofEscherichia coli is RpoS regulated suggesting that it plays an important role during stationary phase growth. The specific role of the gab operon is metabolism of y-aminobutyrate (GABA). GAB~ a four carbon amino acid, is the product ofthe glutamate decarboxylase reaction in E. coli and can serve as a source of carbon and nitrogen for members ofthe Enterobacteriaceae.

Interesting]y, we found that we could isolate GABA utilizing mutants even when they carried mutations that were predicted to abolish gab function. However, we were only able to isolate GABA utilizing mutants in the rpos*" background. This suggests two additional features of GABA metabolism in E. coli. One, that the gab operon in

Escherichia coli is indeed dispensable for GABA utilization. Secondly, other RpoS regulated functions, independent ofgab-encoded functions, may allow the cell to metabolize GABA Finally, we have also investigated the inducing effects of GABA on gab expression in minimal and rich media and the role ofglutamate in osmo-tolerance and acid adaptation.

iii ACKNOWLEDGMENTS

Firstly, I would like to thank my supervisor, Dr. H.E. Schellhorn for providing me with the opportunity to study at McMaster University in pursuing a M.Sc. degree in his laboratory. I am grateful to him for his constructive criticism, patience, and overall support throughout the course of this project. I am also grateful to Dr. Dick Morton for his comments and suggestions during the thesis write-up process and for sitting on my committee.

My sincere thanks to my fellow lab members, both past and present, namely

Jonathon P. Audia, Chantalle K. D' Souza, Lily Chang, Galen Chen, Dr. Rouha Kasra and ofcourse all the summer students for all their technical advice, support and the laughs. Many thanks to all the friends I have made in the Biology Department. All this has helped make my stay here at McMaster University an enjoyable and memorable one.

Finally, I would like to thank my parents for their unconditional love and their continued support in every way possible. Also, I would like to thank God for putting me in the position to make something out of my life.

iv TABLE OF CONTENTS Page

Abstract iii

Acknowledgments iv

Table of Contents v

List of Abbreviations Vlll

List ofTables ix

List of Figures X

CHAPTER!.

INTRODUCTION.

1.1 Overview 1

I .2 RpoS as a regulator 2

1.3 GABA: an inhibitory neurotransmitter in eukaryotes 4

1.4 The Gab Operon 5 1.4.1 Organization of the gab cluster 5 1.4.2 Amino acid homologies of the gab gene products 7

1.5 Pathways Involved in GABA utilization 9 1.5.1 GABA transport in E. coli 9 1.5.2 GABAmetabolisminE. coli 10

1.6 Response of the gab operon to catabolite repression 12

I.7 Pathway of arginine catabolism 15

1.8 Project Outlines and Objectives 15

v CHAPTER2.

MATERIALS AND METHODS.

2.1 Bacterial Strains 24

2.2 Chemicals and Media 24

2.3 Growth Conditions 24

2.4 Assay for (3-galactosidase 24

2.5 PCR amplification and sequence analysis 25

2.6 Isolation of GABA utilizing mutants 26

2. 7 Genetic methods 28

2.8 Isolation ofGABA utilizing mutants from fusion strains 28

2.9 Calculation of mutational frequency 29

2.10 Preparation of Conditioned Medium 29

2.11 GABA induction ofgab expression in Minimal Medium 30

2.12 Acid Adaptation Test 30

2.13 Osmoprotection Test 31

CHAPTER3.

RESULTS

3.1 GABA utilizing mutants obtained 37

3.2 (3-galactosidase assay using transduced GABA mutants 38

3.3 Sequence analysis of the gab promoter in the GABA mutants 39

3.4 Dispensability of the gab operon for GABA utilization 40

vi 3.5 Putative inducers ofgab expression 42 3.5.1 Conditioned Medium 42 3.5.2 LB media supplemented with GABA (0.2%) 43 3.5.3 Minimal media supplemented with GABA (0.2%) 44

3.6 Role of glutamate in acid adaptation and osmotolerance 44

CHAPTER4.

DISCUSSION 79

4.1 References 87

vii List of Abbreviations

ADC Arginine decarboxylase-initiated pathway AST Arginine succinyltransferase pathway ea2+ Calcium csiD Carbon-starvation inducible gene D CNS Central nervous system cAMP-CRP cyclic adenosine monophosphate receptor protein oc Degrees Celsius DNA Deoxyribonucleic acid dNTP 2'-deoxynucleotide-5' -triphosphate GABA y-aminobutyrate Group I (CN) Carbon and nitrogen utilizing GABA mutant Group II (N) Nitrogen utilizing GABA mutant Group III (C) Carbon utilizing GABA mutant LB Luria-Bertani LXS Luria-Bertani-5-bromo-4-chloro-3-indolyl-P-D-galactopyranoside­ Streptomycin J.Lg Microgram J.Ll Micro litre mg Milligram ml Milliliter mM Millimolar J.Lm Micro molar MF Mutational frequency ng Nanogram nm Nanometer OD Optical density 0/N Overnight ONPG o-nitrophenyl-P-D-galactopyranoside PCR Polymerase chain reaction REP Repetitive extragenic palindromic elements rpm Revolutions per minute RNA Ribonucleic acid uv Ultraviolet X-Gal 5-bromo-4-chloro-3-indolyl-P-D-galactopyranoside

viii List of Tables

Page CHAPTER I

Table I. Members of the RpoS regulon in E. coli. 17

CHAPTER3

Table 2a. E. coli strains and bacteriophage used in this study. 47

Table 2b. E. coli strains and bacteriophage used in this study. 49

Table 3. Analysis ofGABA mutants obtained for utilization ofGABA as a carbon and/or nitrogen source. 50

Table 4. GABA mutants obtained from wild type fusion backgrounds. 51

Table 5. Over-expressing GABA mutants obtained. 52

Table 6. Expression ofgab genes in response to conditioned medium. 53

Table 7. Expression of gab genes in response to GABA (0.2%). 54

Table 8. Growth of wild type and GABA mutant strains on various weak acids in the presence and absence ofGABA 55

Table 9. Growth of wild type and GABA mutant strains on varying concentrations of salt in the presence and absence of glutamate. 56

ix List of Figures

Page

CHAPTER!

Figure 1. Schematic representation of the gab region. 18

Figure 2. The pathway of GABA degradation in E. coli. 20

Figure 3. The arginine decarboxylase-initiated (ADC) pathway of arginine catabolism. 22

CHAPTER2

Figure 4. Strategy employed to isolate three different groups of GABA utilizing mutant strains directly from the wild type (MC4100) background. 33

Figure 5. Schematic representation of the strategy used to isolate GABA utilizing mutants directly from strains carrying lacZ fusions to members of the gab operon. 35

CHAPTER3

Figure 6a. Expression ofgabP in wild type strains 57 Figure 6b. Expression ofgabP in CN GABA utilizing mutant strains 59 Figure 6c. Expression of gabP inN GABA utilizing mutant strains 61

Figure 6d. Expression ofgabP in C GABA utilizing mutant strains 63

Figure 7. Clustal W sequence alignment results ofPCR amplified csiD promoter region in wild type (MC41 00) strain and three classes of GABA utilizing mutant strains generated from MC41 00. 65

Figure 8. Replica plate pattern of wild type strain HSI057° 0 (rpoS, gabD-lacZ), its rpoS derivative HS I 057p , and Group I (CN) GABA utilizing mutants (HS1903a-c) generated from the wild type rpoS fusion background only. 67

X Figure 9. Expression of gabD in wild type and "over-expressing" GABA utilizing mutant strains. 69

Figure 10. Conditioned medium as an inducer of gab expression. 71

Figure 11. GABA as an inducer ofgab expression in LB media 74

Figure 12. GABA as an inducer ofgab expression in minimal media 77

XI CHAPTERI. llffRODUCTION

1.1 Overview Adaptation to different environmental stresses is critical for the survival of microorganisms. In their natural habitats, bacteria frequently encounter suboptimal growth conditions and in order to cope with these unfavourable circumstances, most microbes have developed complex stress response mechanisms. For instance, gram positive bacteria such as Bacillus subtilis respond to stresses by differentiating into highly resistant spores, while gram-negative species such as Escherichia coli and Salmonella typhimurium make use of stationary phase response mechanisms (Hengge-Aronis,

1996b).

Stationary phase cultures are easily distinguishable from those existing in exponential phase growth. This is because cells go from a state ofmaximal cell division and proliferation in exponential phase (generation time for E. coli in LB media is 25 minutes) to one of minimal cell growth and dormancy associated with stationary phase.

Interestingly, this shift from exponential to stationary phase is accompanied by many changes in cellular physiology and morphology ofthe bacterial cultures with the purpose to offer longevity and increased resistance to the stresses associated with post­ exponential phase growth. For instance, E. coli cells, normally rod-shaped during exponential phase growth, assume a smaller and more spherical shape in stationary phase

(Lange et al., 1991a; Reeve et al., 1984). Moreover, the cytoplasm of the cells 2 condenses, the volume ofthe periplasm increases (Lange et al., 1991a; Reeve et al.,

1984), the cell wall becomes cross linked and there is an increase in the degree ofDNA supercoiling (Balke et aL, 1987).

1.2 RpoS as a regulator

The end of exponential phase is marked by increased expression of genes which seem to help bacteria combat multiple stresses. Expression of these stationary phase genes is under the control of the rpoS encoded c:l subunit (also called a38 or RpoS)

(Lange et al., 1991b; Mulvey et aL, 1989) .This sigmaS subunit ofRNA polymerase is a master regulator that recognizes a set of promoters poorly recognized by the

10 "housekeeping sigma factor" a •

The RpoS regulon is quite large. Over forty genes and/or operons encoding a diversity of functions are known to be regulated by c:l (Summarized in Table 1). These functions include DNA protection and repair (encoded by katE, katG, aidB and xthA)

(Hengge-Aronis, 1996b), thermotolerance (otsBA, htrE) (Kaasen et al., 1992; Hengge­

Aronis et aL, 1991), osmotolerance (osmY, proP, otsBA) (Lange et aL, 1993; Manna et aL, 1994; Kaasen et aL, 1992; Hengge-Aronis et aL, 1991) and glycogen synthesis (gigS)

(Hengge-Aronis et aL, 1992). In some cases, RpoS also contributes to the virulence and pathogenesis of bacteria. RpoS is important for acid tolerance in Salmonella typhimurium (Lee et aL, 1995) and in Shigella.flexineri (Small et aL, 1994). Likewise,

RpoS regulates the expression of spvR inS. typhimurium (Kowarz et al., 1994). SpvR activates the expression of the plasmid borne virulence genes (spvABCD) (Kowarz et aL,

1994) which are required for intracellular growth. Furthermore, in Yersinia 3 enterocolitica, an RpoS homologue regulates the enterotoxin gene yst which causes diarrhea (lriarte et al., 1995).

The important role ofRpoS in the E. coli stress response has led to the examination of similar RpoS systems in other bacteria. Homologues of rpoS have been discovered in Salmonella typhimurium (Kowarz et al., 1994), Shigella flexineri (Small et al., 1994), Pseudomonas aeruginosa (Fujita et al., 1994), Yersinia entercolitica (lriarte et al., 1995), Erwinia carotovora (Becker-Hapak et al., 1997) and Vibrio cholerae (Yildiz et al., 1998). Furthermore, sigma filctors that are structurally and functionally homologous to RpoS have been identified in Acetobacter methanolicus, Xanthomonas camprestris,

Pseudomonas putida, and Rhizobium meliloti (Miksch et al., 1995). The role ofRpoS in multiple stress protection seems to be fairly conserved amongst bacteria.

Regulation ofRpoS expression is controlled at the levels of transcription, translation and protein stability (Lange et al., 1994b). While RpoS may be the predominant sigma filctor in stationary phase, RpoS and many RpoS--dependent genes are also induced under various stress conditions including osmotic stress (Lange et al.,

1994b), acid stress (Lee et al., 1995) and heat shock (flshage et al., 1996) during exponential phase growth. Moreover, expression of many RpoS--dependent genes is far more complex than a simple activation by RpoS in that additional regulators are often involved either through binding near promoter regions or through cascade-like regulation.

For example, csiE and csiD are co-regulated by RpoS and cAMP-CRP (Marschall et al.,

1995; Marshall et al., 1998), while osmY (Lange et al., 1993) and osmC (Bouvier et al.,

1998) are co-regulated by the DNA-binding protein Lrp (leucine responsive protein). Fis 4

(Xu et al., 1995), IHF (Lange et al., 1993) and H-NS (Barth et al., 1995) have also been identified as modulators ofRpoS-dependent genes. Needless to say, more research needs to be done to learn more about this global stress response regulator known as RpoS and the complex network of genes it regulates. Doing so will undoubtedly have practical applications in the fields ofpharmacology and medicine.

1.3 GABA: an inhibitory neurotransmitter in eukaryotes

GABA, or y-aminobutyrate, is an amino acid It is a four carbon compound and its chemical structure is as follows: NH2-(CH2)3-COOH. Gamma-aminobutyric acid is the product of the glutamate decarboxylase reaction in E. coli (Marcus et al., 1969). It is also an intermediate in the pathway ofL-arginine and putrescine (Figure 3) catabolism in

E. coli and K aerogenes (Friedrich et al., 1979; Shaibe et al., 1985). GABA can serve as a source of carbon and nitrogen for members of the Enterobacteriaceae and it is likely that small amounts ofthis amino acid are present in the natural environments ofthe members of the Enterobacteriaceae as a result of various degradative processes.

In 1950, GABA was independently identified and reported to be present in the vertebrate brain by Roberts and Frankel (1950) and Awapara et al. (1950). Early electrophysiological work carried out primarily in crustaceans firmly established GABA as an inlnoitory neurotransmitter in invertebrates. By the early 1970s, GABA had been shown to satisfY all of the classical criteria of a neurotransmitter.

Amino acids are amongst the most abundant of all neurotransmitters present within the central nervous system (CNS). The majority of neurons in the mammalian brain utilize either glutamate or GABA as their primary neurotransmitters (Guyton and 5

Hall, 1996). In fact, GABA and glutamate have opposing inhibitory and excitory actions, respectively, and as a result, control to a large degree the overall "balance" in the CNS.

In this way, these two neurotransmitters serve to regulate the excitability of virtually all neurons in the brain and have been implicated as important mediators of brain function or dysfunction (Hertz, 1983).

GABA is the major inhibitory neurotransmitter in the central nervous system. It is estimated that 30-40% of all CNS neurons utilize GABA as their primary neurotransmitter (Hertz, 1983). It is secreted by neurons in the spinal cord, basal ganglia, cerebellum, and many areas of the cortex. Like most neurotransmitters, GABA is stored

2 in synaptic vesicles and is released in a Ca +-dependent manner upon depolarization of the presynaptic membrane. Following release into the synaptic cleft, GABA's actions may be terminated by selective uptake into presynaptic terminals and/or surrounding glia

(the supporting cells of the nervous system that produce myelin sheath) where it gets metabolized. The world ofpharmacology is heavily involved in creating drugs antagonistic to various parts of the GABA degradation pathway. For instance, GABA transaminase inhibitors cause high concentrations of GABA in tissues which is beneficial in lowering anxiety, preventing panic attacks, and protection against epileptic seizures

(Guyton et al., 1996).

1.4 The Gab Operon:

1.4.1 Organization of the gab cluster

The specific role of the gab gene cluster is metabolism ofy-aminobutyrate

(GABA) in Escherichia coli (Dover and Halpern, 1972a, 1972b, 1974). This cluster 6 consists of four genes, one regulatory gene and three structural genes (Figure 1). The three structural genes in this operon are gabD, gabT and gabP, where gabD and gabT encode the two metabolic enzymes, succinic semialdehyde dehydrogenase and glutamate succinic semialdehyde transaminase, respectively, involved in GABA catabolism; gabP codes for the y-aminobutyra.te transporter (Metzer et al., 1979; Niegemann et al., 1993).

The control gene, gabC, specifies a gab repressor and is therefore believed to be a negative regulator ofgab operon expression (Metzer et al., 1990). Previous studies

(Kahane et al., 1978) seem to support this hypothesis where strains carrying the wild-type allele ofgabC exlnbit low expression of the gab genes but gabC constitutive strains showcase a derepression of the synthesis of enzymes involved in GABA utilization. The gab operon has been shown to be highly RpoS dependent (Schellhorn et al., 1998). The map position for the gab loci on the E. coli K-12 chromosome is at 57.5 min (Metzer et al., 1979).

The nucleotide sequences of the structural genes of the gab operon have been determined (Niegemann et al., 1993; Bartsch et al., 1990b). The three structural genes in the gab operon are transcnbed unidirectionally from a common promoter upstream of gabD and their orientation within the cluster is 5'-gabD-gabT-gabP-3' (Figure 1)

(Bartsch et al., 1990b; Metzer et al., 1990; Niegemann et al., 1993).

There lies an intergenic region of 234 base pairs between gabT and gabP, which contains three repetitive extragenic palindromic (REP) sequences with a length of35-bp each (Niegemann et al., 1993). REP sequences ofthis type have been detected in the intergenic regions of a large number of operons from E. coli and Salmonella typhimurium 7 and are believed to pJay a role in regulating expression of genes within these operons

(Gilson et al., 1984; Stern et al., 1984). In fact, REP units are found between genes with a several-fold step-down in expression of the distal gene compared to the promoter proximal gene (Gilson et al., 1984; Stern et al., 1984). These elements are possibly involved in downstream gene regulation within an operon by mechanisms like transcription termination via formation of stable hairpin loops, RNA cleavage or degradation arrest (Niegemann et al., 1993). Nevertheless, differences in reJative expression of the gab genes are also suggested by codon usage. The calculated codon bias index of0.34 and 0.27 for gabD and gabT, respectively, corresponds with the value of medium expressed E. coli genes, while the bias index ofgabP (0.12) indicates a low expression rate of the gene (Niegemann et al., 1993).

1.4.2 Amino acid homologies of the gab gene products

The gabD gene consists of 1,449 nucleotides encoding a protein of 482 amino acids with a molecuJar mass of 51.7 kDa (Niegemann et al., 1993). The product ofgabD, succinic semialdehyde dehydrogenase, shows significant homologies to the NAD+­ dependent aldehyde dehydrogenase fromAspergillus nidulans (Pickett et al., 1987) and several mammals (Hsu et al., 1988; Farres et al., 1989) and to the tumor-associated

NADP+ -dependent aldehyde dehydrogenase from rats (Jones et al., 1988). The amino acid identity score was 35-38% between succinic semialdehyde dehydrogenase and the

NAD+ -dependent aldehyde dehydrogenases and 29% between succinic semialdehyde dehydrogenase and the NADP+ -dependent tumor associated aldehyde dehydrogenase. 8

The gabT gene consists of 1,281 nucleotides encoding a 426 amino acid protein with a calculated molecular mass of 45.76 kDa (Bartsch et al., 1990b). The product of gabT, GABA transaminase, shows significant homologies to the ornithine transaminases

(OAT) from Saccharomyces cerevisiae (SCOAT), from rat and from human mitochondria. The amino acid identity score was 28.2% between GABA transaminase and OAT from human mitochondria, 27.9% between GABA transaminase and OAT from rat mitochondria, and 27.5% between GABA transaminase and SCOAT (Bartsch et al.,

1990b). These findings are in accordance with the nucleotide sequence for S. cerevisiae

GABA transaminase (Andre et al., 1990). The tact that ornithine is accepted as a substrate by several GABA transaminases (Schulz et al., 1990; Yonaha et al., 1985;

Yonaha et al., 1980), suggests that the structural relatedness ofthese aminotransferases may reflect functional similarities of these proteins.

The gabP gene consists of 1,401 nucleotides encoding a highly hydrophobic protein of466 amino acids with a molecular mass of51.1 kDa (Niegemann et al., 1993).

The polypeptide consists of 12 highly hydrophobic regions of20 amino acids each, connected by short hydrophilic sequences. TheN- and C-terminal ends of the protein (15 amino acids each) are hydrophilic (Niegemann et al., 1993). The gabP gene product,

GABA permease, shows strong resemblance to the proline, arginine and histidine permeases from Saccharomyces cerevisiae (Hoffinann, 1985; Tanaka et al., 1985;

Vandenbol et al., 1989) with amino acid identity scores of27.5%, 30.4% and 28%, respectively. The GABA permease also shows significant homology to the proline transport protein from Aspergillus nidulans (Sophianopoulou et al., 1989) with an amino q acid identity score of31.2%. The highest degree ofhomology, however, was observed with the aromatic amino acid transport protein from E. coli, the product ofthe aroP gene

(Honore et al., 1990). Furthermore, no sequence similarities were found between the

GABA transporter and the two glutamate carriers from E. coli encoded by the gltS and gltP genes, respectively (Wallace et al., 1990; Kalman et al., 1991).

1.5 Pathways Involved in GADA utilization:

1.5.1 GADA transport in E. coli

Since GABA may serve as a source of either carbon or nitrogen, it is advantageous to the cell to be able to enhance its uptake in response to signals which relay a shortage in either one ofthem. To serve this purpose, there exists in E. coli a highly specific membrane transport carrier for GABA encoded for by the gabP gene.

GABA permease, encoded by gabP, is responsible for the transport of y-aminobutyrate

(GABA) across the E. coli cell membrane. The permease is active in whole cells as well as in rightside-out vesicles (Kahane et al., 1978) and uptake of GABA is driven by membrane potential but abolished by proton ionophores (Niegemann et al., 1993).

Previous studies (Kahane et al., 1978; Niegemann et al., 1993) show that the transporter is highly specific for y-aminobutyrate transport, being unable to recognize alternative ligands such as the 20 common a-amino acids, ornithine and putrescine, all of which exhibit superficial structural similarity to GABA The requirement ofthe GABA transport system has been demonstrated by the use of GABA transport mutants (Kahane et al., 1978) where mutants isolated were found to have lost the ability to accumulate

GABA due to mutational lesions existing within gabP, but retained the parental high 10 levels of GABA transaminase and succinylsemialdehyde dehydrogenase, the degradative enzymes of the GABA catabolic pathway.

More recent studies (Brechtel et al.,l995a; King et al., 1995b), however, now

indicate that the gabP transport channel can recognize and translocate a far more diverse

range of chemicals than previously imagined. In fact, numerous molecules having

different sizes and shapes are capable of interacting with the GABA permease either as

inhibitory ligands and/or as transported substrates. Moreover, there is evidence which

suggests that these inhibitory molecules exert their affect by acting within the gabP

transport channel itself rather than at an allosteric inhibitory site (Brechtel et al., 1996).

1.5.2 GABA metabolism in E. coli

Once transported into the E. coli cell, y-aminobutyrate catabolism occurs in two

successive steps leading to its conversion to succinate which then gets utilized in the

TCA cycle (Figure 2). The enzymes involved in this two step degradative reaction are 4-

aminobutyrate transaminase, encoded by gabT, and succinic semialdehyde

dehydrogenase, encoded by gabD, respectively.· The pathway of GABA breakdown in E.

coli K-12 seems to be the same as that in Pseudomonas jluorescens (Scott et al., 1959;

Jakoby et al., 1960; Jakoby et al., 1959). The first step is transamination between GABA

and a-ketogluterate to yield succinyl semialdehyde and glutamate. This reaction is

catalyzed by y-aminobutyrate-a-ketoglutarate transaminase. Succinyl semialdehyde is

then oxidized to succinate by succinyl semialdehyde dehydrogenase for use in the

tricarboxylic acid cycle. II

Wild-type E. coli K-12 strains unable to utilize GABA as the sole source of nitrogen or carbon or both exhibit very low activity of these two enzymes involved in

GABA breakdown, but mutants selected for their ability to utilize GABA as the sole source of nitrogen show a six to nine-fold increase in the activities of these two enzymes

(Dover and Halpern, 1972a). This increase in enzymatic activity is not accompanied by any changes in the affinities ofthe mutant enzymes for their respective substrates (a­ ketoglutarate for gabT and NADP for gab D) (Dover et al., 1972b) which strongly suggests that the mutations responsible which enable the cells to utilize GABA as the sole source of nitrogen occur at a control locus which regulates the synthesis of these enzymes, rather than affecting a structural gene. A similar situation has been descn"bed on the utilization ofhistidine by Salmonella typhimurium (Smith et al., 197la; Smith et al., 197lb). The wild-type strain unable to utilize histidine as a single source of carbon or nitrogen or both, exlnoits very low activities ofhistidase and urocanase, enzymes which catalyze the first two steps in the breakdown ofhistidine to glutamate. However, a mutant isolated for its ability to utilize histidine as a nitrogen source showed greatly increased levels of these two enzymes.

Further support for the hypothesis that the two enzymes of the pathway ofGABA utilization belong to a single regulatory unit is evident by the fact that the synthesis of y­ aminobutyrate transaminase and succinic semialdehyde dehydrogenase is strictly coordinate under a great variety of growth conditions (Dover and Halpern, 1972b). That

is, changes in conditions affect the rate of synthesis ofthese two enzymes to the same extent. These observations disprove earlier evidence which argued against the two 12

GABA enzymes constituting a single operon (Dover and Halpern, 1974) and support the contention that the two enzymes of the GABA utilization pathway constitute an operon

(Jacob et al., 1961; Bartsch et al., 1990b; Niegemann et al., 1993).

1.6 Response of the gab operon to catabolite repression

There is significant evidence which suggests that the gab operon is sensitive to strong catabolite repression. Studies (Kahane et al., 1978) suggest that the synthesis of the GABA transport carrier is subject to strong catabolite repression. Evidence for this springs from the fact that the rate ofGABA uptake by cells grown in glucose-NH.t + medium is only about one-ninth of the rate obtained with succinate-NH.t..-grown cells which in turn is 50% less than the GABA transport activity of glucose-GABA grown cells (Kahane et al., 1978).

Likewise, the two enzymes involved in the GABA degradation pathway in E. coli

K-12 are also subject to severe catabolite repression (Magasanik et al., 1961). In cultures grown in minimal media with ammonium salts as the sole source of nitrogen, both enzymes are severely repressed by glucose. Under these conditions, whether serving as the sole carbon source or when present in the medium together with succinate or GABA, glucose repressed the synthesis of y-aminobutyrate transaminase and succinic semialdehyde dehydrogenase to less than 10% of that found in cells grown in succinate medium in the absence of glucose. Addition of GABA did not relieve glucose repression, nor did GABA affect the levels of the two enzymes in succinate-ammonia medium in the absence of glucose (Dover and Halpern, 1972b). l3

Interestingly, unlike other catabolite-sensitive systems, the GABA system resists catabolite repression under conditions of limited nitrogen supply (Dover and Halpern,

1972b). That is, escape from repression by glucose occurs only when the ammonia in the growth medium is substituted with GABA as the sole nitrogen source. However, this escape from repression is not a specific effect of GABA because this escape from repression of the enzymes involved in GABA metabolism may also be brought about by substitution of ammonia with glutamate or aspartate, although the extent of derepression is less marked than in the case of GABA as the sole nitrogen source (Dover and Halpern,

1972b). Nevertheless, in E. coli, GABA nitrogen utilizing mutants have been isolated in which the synthesis of y-aminobutyrate succinic semialdehyde transaminase remains repressed in glucose media even under conditions ofnitrogen limitation, while the synthesis of succinic semialdehyde dehydrogenase is derepressed under the same conditions (Dover and Halpern, 1974). This finding seemed to argue that the transcription of the two structural genes involved in GABA metabolism is initiated at two separate promoters and that there is more than one dehydrogenase. Metzer and Halpern

(1990) cloned the E. coli K-12 gab region and postulated that the gab genes may be divergently transcn"bed by two different promoters. Nevertheless, more recent studies now confirm that the two structural genes of the GABA regulon do constitute one operon

(Niegemann et al., 1993).

A similar escape from catabolite repression has been described for the histidine utilization enzymes in Aerobacter aerogenes following substitution of the ammonia in the growth medium with histidine (Neidhardt et al., 1957). Conversely, in Salmonella 14 typhimurium, in which histidine utilization is also subject to catabolite repression, no such escape was observed (Magasanik et al., 1961). It is important to mention that this escape from glucose repression upon substitution of ammonia with other nitrogen sources was found to be specific for the enzymes of the GABA utilization pathway and did not

involve other enzymes sensitive to catabolite repression, such as p-galactosidase and aspartase. In fact, P-galactosidase and aspartase were as strongly repressed by glucose in

the presence of GABA or aspartate as in the presence of ammonia as the source of nitrogen (Dover and Halpern, 1972b).

It has also been shown that the factor responsible for this relief from repression is glutamine synthetase which is derepressed under conditions of nitrogen limitation

(Zaboura et al., 1978). Experiments with glutamine synthetase negative and glutamine

synthetase constitutive mutants strongly indicate that glutamine synthetase is the effector

which regulates GABA carrier synthesis as a function of nitrogen availability (Zaboura et al., 1978) and by doing so acts as a positive regulator in the E. coli GABA controL

Glutamine synthetase probably acts as an activator of transcription in the gab system of

E. coli (Magasanik. et aL, 1974). Activation by glutamine synthetase is not required in

situations where no catabolite repression prevails, probably because ofthe availability of

sufficient CRP-cyclic AMP complex to serve as an alternative activator. Another

transport system that may be regulated by glutamine synthetase is that of glutamine in

Salmonella typhimurium (Ayling et al., 1976) and possibly also in E. coli (Furlong et al.,

1975). 15

1.7 Pathway of arginine catabolism

L-Arginine can serve as sole carbon and nitrogen source forK aerogenes, as a poor nitrogen source for S. typhimurium and E. coli, and as a source of carbon for some E. coli and Proteus strains (Cunin et al., 1986; Friedrich et al., 1978; Kustu et al., 1979; Shaibe et al., 1985). In bacteria, two pathways exist for arginine catabolism (Cunin et al., 1986).

The first pathway (Figure 3) is initiated by arginine decarboxylase, hence the name ADC pathway, which degrades arginine to agmatine and then eventually through a series of steps to succinate (Cunin et al., 1986). This pathway (Shaibe et al., 1985; Wilson et al.,

1969) consists of six enzymatic steps and it converges with the GABA degradation pathway (Figure 2). Another pathway of arginine catabolism, called the arginine succinyltransferase (AST) pathway, exists in bacteria which is initiated as a response to succinylation of arginine. Degradation of arginine via this AST pathway eventually yields glutamate and succinate (Cunin et al., 1986).

1.8 Project Outlines and Objectives

This project was initiated by Dr. Rouha Kasra and Dr. Herb Schellhorn in the summer of 1999. Dr. Kasra began by isolating the three classes of GABA utilizing mutants from the wild type MC4100 background under Dr. Schellhorn's supervision. Due to problems in locating her strains after her departure, this work was repeated by myself in the fall of

1999, shortly after being assigned this project. Dr. Schellhorn carried out the transductions referred to in Materials and Methods. 16

The goal of the project was to characterize GABA mutants in order to elucidate the mutations responsible for GABA utilization in these mutant strains. Primers were designed to amplify the potential gab promoter (csiD promoter) and sequence analysis was performed to identify the possible mutation site. Furthermore, as a result of this study, we were able to show the existence of alternative GABA utilization pathway(s) that do not require the gab operon. Finally, we tested various putative inducers ofgab expression and analyzed the protective functions of glutamate against salt and acid stress. 17

Table 1. Members of the RpoS Regulon in E. co/i.8

Gene/ RpoS- Operon Function dependenceb Reference aidB involved in leucine metabolism; defends 20 Landini et a/.,1984; against DNA methylated damage Volkert eta/., 1994 appY controls expression of hya and cyxAB- appA 2 Atlung et a/. ,1994 bolA confers round morphology, controls synthesis of PBP6 12 Aldea et a/. ,1989 cbpA chaperone, curved DNA binding protein 5 Yamashino et a/.,1994 cfa ? 6 Wang eta/. ,1994 csiE ? 15 Weichert eta/., 1993 fie involved in folate synthesis 3 Utsumi et a/.,1993 gabT involved in GABA metabolism, encodes 18 Beca-Delancey et glutamate-succinic semlaldehyde transaminase a/., 1999 gabD involved in GABA metabolism, encodes succinic semialdehyde dehydrogenase 19 Schellhorn eta/., 1998 gabP involved In GABA transport, encodes GABA 31 Schellhorn et a/.,1998 permease gigS glycogen synthesis 3 Hengge-Aronis and Fischer, 1992 himA gene regulation 2 Aviv eta/., 1994 hmp anaerobic metabolism of nitrogen 2 Membrillo- Hernandez eta/., 1997 hyaABC DEF oxidation of hydrogen 20 Brondsted and Atlung, 1994 katE 52 Sak eta/., 1989; Schellhorn and Stones, 1992 /deC maintains pH homeostasis 7 Kikuchi eta/., 1998; Van Dyk et a/., 1998 osmY ? 68 Lange and Hengge­ Aronis, 1993 otsBA trehalose synthesis, osmoprotection, 32 Kaasen et a/., thermotolerance 1992; Hengge­ Aronis eta/., 1991 poxB synthesis of acetate 99 Chang et a/., 1994 proP osmoprotection 4 Manna et a/., 1994 xthA DNA repair, hydrogen peroxide resistance 3 Sak et al, 1989 yciG ? 188 Van Dyk et a/., 1998 yohF ? 51 Fang eta/., 1996; Van Dyk et a/., 1998 1AJI genes/operons listed are positively regulated by RpoS. ~alues for RpoS dependence were calculated by comparing the maximum activity of each fusion in an rpoS+ and rpoS" background. 18

Figure 1. Schematic representation of the gab region. Shown are the structural genes of the gab cluster, the nucleotide sequence and important structural features (transcriptional and translational start sites, -10 and -35 positions of the csiD promoter, and highly conserved nucleotides in the CRP binding site) of the csiD promoter and the primers designed to amplify the csiD promoter. Note that gabC is located upstream of the gab operon (Diab, K., Metzer, E. and Halpern, Y.S. unpublished, U68244). 19

1.0 kb

5'-AGGGAAAACAGGAGAACCGC-3'

3'-TIAGCAAAGACGCAAAAGCC-5' PCR Product=573 nt t

57.5 min.

po-10

REP element

csiD promoter

gctttagttattcatttcctgtctcactttgccttaataccctacgttaaatgttactaatttgttgcttttgatcacaataagaaaacaa

tatgtcgcttttg~tttttcagaaatgtagatatttttagatt~tggctacgaaatgagcatcgccatgtcaccctacatctca -35 -10 +1 _..,. taagaggatcgcttctg~ 20

Figure 2. The pathway of GABA degradation in E. coli. Shown are the three gab genes and the enzymes that they encode which are involved in the metabolism of GABA. The first step in this process is the transport of GABA across the E. coli cell membrane. This step is catalyzed by the product of gabP. The next two steps are responsible for GABA catabolism: the first of these being a transamination reaction catalyzed by the product of gab T, and the last being an oxidation reaction catalyzed by the gabD gene product. 21

y-am inobutyric acid (GABA) NH2-(CH2)a-COOH

E. coli cell membrane------GABA permease (gabP) GABA

a -ketoglutarate Glutamate-succinic semialdehyde transaminase glutamate (gabT)

O=C-H I (CH2h Succinic semialdehyde I COOH

NAD(P) Succinic semialdehyde NAD(P)H dehydrogenase (gabD)

Succinic acid COOH- (CH2h -COOH 22

Figure 3. The arginine decarboxylase-initiated (ADC) pathway of arginine catabolism. Enzymes involved at every step are indicated and genes encoding for these respective enzymes are italicized in parenthesis (Shaibe et al., 1985). 23

NH2-C=NH I arginine (CH2)s I COOH-CH-NH2 kginine decarboxylase (speA)

~ -co2 ornithine 2 NH2-(CH2)s-CH-NH2 NH -r=NH agmatine I NH-(CH2)4-NH2 COOH ~-.-t~O ur~gmatine ornithine 2 ureohydrolase decarboxylase (speS) (speC) NH2-(CH2)4 NH2 putrescine

a.-ketoglutarate>! putrescine transaminase (pat) glutamate

NH2-(CH2)s-CH=O y-amlnobutyraldehyde NAD>{ NADH pyrroline dehydrogenase (prr)

NH2-(CH2)s-COOH

GABA CHAPTER 2. MATERIALS AND METHODS

2.1 Bacterial strains

The bacterial strains used in this study are listed in Tables 2a and 2b.

2.2 Chemicals and media

All chemicals were supplied by either Fisher Scientific, Ltd. (Toronto, Ontario,

Canada), Sigma Chemical Co. (St. Louis, Mo.), Gibco BRL (Burlington, Ontario,

Canada), or BDH Inc. (Toronto, Ontario, Canada). Antibiotics and other nonautoclavable

stock solutions were filter sterilized with Gelman Sciences (Ann Arbor, Michigan)

acrodisc sterile filters (pore size, 0.45 urn). Liquid media was prepared as descnbed by

Miller (1992). The concentrations of antibiotics used were as follows: kanamycin (50

ug/ml), streptomycin (100 ug/ml), and tetracycline (15 ug!ml). X-Gal (5-bromo-4- chloro-3-indolyl-13-D-galactopyranoside) was used at a concentration of 50 ug!ml.

2.3 Growth Conditions

All strains used were MC4100 derivatives (see Table 2a). Cultures were grown

routinely in Luria-Bertani (LB) rich media at Jrc containing the appropriate antibiotics.

Cell growth was monitored spectrophotometrically (UV-VIS spectrophotometer, model

UV-1201, Shimadzu Corporation, Kyoto, Japan) at optical density of600 nm (00600).

2.4 Assay for 13-galactosidase

13-galactosidase activity was assayed using o-nitrophenyl-13-D-galactopyranoside

(ONPG) as the substrate and activity is expressed in Miller units (Miller, 1992).

24 25

Cultures were grown overnight in LB media containing the appropriate antibiotics. For expression studies, the bacterial cultures were maintained in early exponential phase (OD600 < 0.2) for at least eight generations prior to the start of each growth experiment. This ensured that measured 13-galactosidase activity in early exponential phase cultures was due to de novo enzyme synthesis rather than being due to carry over from overnight stationary phase cultures. To ensure that the bacteria were in exponential growth, overnight cultures were diluted 1:1000 in fresh LB medium and incubated at 3 rc with vigorous shaking (250 rpm). After 3 hours, the cultures were diluted again 1:100 (00600=0.02). For induction studies, conditioned medium and LB media supplemented with GABA (0.2%) were used as the growth medium. For the latter, the assay was done in duplicate. Subsequently, the cells were sampled and assayed for 13- galactosidase activity at the times indicated. Cultures were grown in flasks at a culture/flask volume ratio of 1 to 5 at 3'?C and 250 rpm to maintain proper aeration. 13- galactosidase activity was assayed as descnbed by Miller (1992). Units of activity were calculated as [1000 X 004zo]/[time of incubation (min.) X volume (ml) X 00600] and were expressed as Miller units.

2.5 PCR amplification and sequence analysis

Primers used to amplify the E. coli csiD (carbon starvation-inducible gene) promoter (Marschall et al., 1998) were as follows: forward primer (AB19298) 5'­

AGGGAAAACAGGAGAACCGC-3' and reverse primer (AB19299) 5'­

CCGAAAACGCAGAAACGATT-3'. These primers were specifically designed using

Primer Design Program (http://www.williamstone.com/) on the GenBank database 26 sequence (AE000351) consisting ofthe csiD promoter sequence as listed in Marschall et al. (1998). Each PCR tube contained 1 X PCR buffer (50 mM KCl, 20 mM Tris pH 8.4),

50 pmoles of each of the forward and reverse primers, 0.4 mM of each of the 4 nucleotide triphosphates, 4 mM MgCh,- 50 ng of E. coli MC4100 template DNA and- 5 U ofTaq polymerase in a final volume of 50 ul. Amplification was performed under the following conditions: (1) 95°C for 2 minutes (pre-denaturation step); (2) (denaturation: 95°C for 15

seconds, annealing: 62°C for 30 seconds, elongation: 72°C for 1 minute, 25 cycles); (3)

72°C for 7 minutes (final extension step). We obtained a DNA fragment of573 bp in length consisting of the csiD promoter region. PCR products were purified using a

QIAquick PCR Purification Kit (Qiagen Inc. Valencia, CA). All PCR-amplified products

were sent for DNA sequencing at the Mobix facility (McMaster University) and sequence comparison of amplified sequences to the wild type (MC41 00) csiD promoter sequence

was done using Clustal W (1.8) multiple sequence alignment program

(http://www2.ebi.ac. ukl).

2.6 Isolation of GADA utilizing mutants

Many strains ofE. coli K-12 (such as MC4100) cannot utilize GABA, except as a nitrogen source in the presence of a poor carbon source such as succinate (Donnelly and

Cooper, 1981). The reason for inability to grow on GABA for these strains is not clear.

However, it is possible to isolate GABA utilizing mutant strains which can use GABA as

sole carbon and/or nitrogen source (Dover and Halpern, 1972a). In this study, three

classes of GABA utilizing mutants, eight mutants of each class, were isolated directly

from the wild type MC41 00 background by selection on minimal media supplemented 27 with y-aminobutyrate (Figure 4). Obtaining these GABA utilizing mutants was to be the first step towards characterizing the factors enabling these strains to metabolize GABA.

Group I mutants (HS 1700) were selected to use GABA as both a carbon and nitrogen source. Group II mutants (HS 1701) were selected to use GABA as a nitrogen source only and Group III mutants (HS1702) could use GABA as a carbon source only.

Group I mutants were selected with minimal salts media (without ~Cl salts)+GABA

(0.7% GABA as nitrogen and carbon source), while Group II mutants were selected with minimal salts media (without ~Cl salts)+glucose+GABA (0.2% glucose as carbon source, 0.2% GABA as nitrogen source). Group III mutants were selected with

M9+~Cl+GABA (0.5% GABA as carbon source). Proline, thiamine and MgS04 were added giving final concentrations of0.002%, 0.001% and 1.0 mM, respectively.

The mutant isolation scheme was carried out in the following manner. Wt.ld-type

MC4100 E. coli cells were streaked out onto aLB plate and incubated overnight at 37"C.

Independent colonies were used to inoculate eight different LB broth tubes which were grown overnight at 37"C to stationary phase. Each ofthese tubes was then used to streak onto an LB-X-gal(5 bromo 4-chloro-3-indolyl-B-D-galactopyronoside)-Streptomycin

(LXS) plate containing streptomycin (50 ug/ml). These eight LXS plates were incubated overnight at 37°C. The purpose of this step was to consolidate phenotypic similarities between Group I and Group III mutants at the genotypic level. Independent colonies from each of the eight LXS plates were then spread plated onto the eight different plates for each of the three types of selective media. These spread plates were incubated at

37°C for 3-4 days. At this point, spontaneous mutants, easily distinguishable from wild- 28 type cells due to the significantly larger colony size of the GABA utilizing strains, were streaked onto fresh selective media to purify the GABA utilizing mutants. After three rounds of purification, these mutant cells were stored (in duplicate) in glycerol at -80°C.

Finally, the phenotype of these GABA utilizing mutants was confirmed by replica plating onto selective media supplemented with GAB A. See Table 2a for complete list of all strains and their genotypes.

2. 7 Genetic methods

After selecting the three classes of GABA utilizing mutants (Table 2a), transductants were made with these mutants as recipients. The genetic procedures for transduction were done by Dr. Herb Schellhorn as described by Silhavy et al. (1984).

LacZ fusions to genes in the gab operon were introduced into the GABA utilizing mutant strains by P !-mediated transduction. P 1 (HS 101 0) was known to contain lacZ fusion to the gabP promoter and P 1 (HS 1057) was known to contain lacZ fusion to the gabD promoter.

2.8 Isolation of GABA utilizing mutants from fusion strains

Results obtained from this study gave reason for postulating an alternate pathway of GABA utilization that may work independently of gab-encoded functions. Therefore, we wanted to test the notion whether the gab operon was dispensable (i.e., non-essential) for GABA utilization. For this purpose, we attempted to isolate GABA utilizing mutants

directly from strains carrying lacZ fusions to genes in the gab operon (HS 1010°,

0 0 HS1010p , HS1057°, HS1057p ) (from Table 2b). MC4100 and HS1600 (from Table 2a)

were used as control (i.e., non-fusion) background strains for isolation of GABA utilizing 29 mutants. The procedure used is outlined in Figure 5. Cells from HS1010° (gabP-lacZ),

0 0 its rpoS derivative HS 101 Op , HS 1057° (gabD-lacZ), its rpoS derivative HS 1057p , and the control non-fusion background strains (MC4100, HS1600) were washed in 10 mM

K2HP04 and 200 ul of each suspension was spread plated onto the three types of selective media plates for the selection of the three classes ofGABA utilizing mutants

(composition of the selective plates same as mentioned previously). The spread plates were incubated at 3~C for 2-3 days. Five mutant colonies of each class generated from the fusion strains were streaked onto LB-X -gal plates and only those mutants that appeared to be over-expressing the functions of the gab operon (by comparing expression levels with the parental fusion strains) were stocked (in duplicate) in glycerol at -80°C.

See Table 5 for list of over-expressing GABA mutants obtained.

2.9 Calculation of mutational frequency

Mutational frequencies for the GABA utilizing mutants obtained from the two rpos+" fusion backgrounds were calculated as follows: cells from the two parental fusion strains were resuspended in 10 mM K2HP04 to an 00600 - 0.2. This corresponds to - 1

X 108 cells/mi. Since 200 ul of cell suspension was spread plated, the total number of cells per plate amounts to - 2 X 107 cells. Therefore, to calculate per cell mutational frequency, the total number of mutant colonies obtained was divided by 2 X 107 cells.

2.10 Preparation of Conditioned Medium

Conditioned medium was prepared as described (Baca-DeLancey et al., 1999).

800 ml ofLB broth was inoculated with 1:6000 dilution of an overnight culture of E. coli

MC4100 followed by shaking (250 rpm) at 37°C. Cells were harvested 3-4 hours after 30 entering stationary phase at an OD6oo of 1.6-2, and cells were pelleted by centrifugation for 10 minutes. The resulting supernatant was supplemented with the addition of20 X

TY (tryptone, yeast extract) to a final concentration of0.5X, and the pH was brought to

7.5. Conditioned medium was then filter-sterilized through a 0.2-um-pore filter and stored at 4°C.

2.11 GADA induction of gab expression in Minimal Medium

MC4100, HS1600 (Table 2a), and all strains listed in Table 2b were grown overnight in LB media with the appropriate antibiotics. Overnight cultures were diluted

1:1000 in fresh LB medium and grown to stationary phase (OD60o ~ 1.0) at 3 rc with shaking (250 rpm). Cells were then pelleted by centrifugation for 10 minutes and washed twice in M9 media (without Nl4Cl salts). This suspension was then replica plated onto minimal salts medium with ammonium as the nitrogen source and either glucose (0.2%) or succinate (0.5%) as the carbon source. Proline, thiamine and MgS04 were added as indicated previously. The test compound (y-aminobutyrate) was added at a concentration of0.2% (Figure 12). The plates were incubated overnight at 3rc.

2.12 Acid Adaptation Test

Glutamate, an amino acid, has been documented to.. serve a protective function against acidic shock (De Biase et al., 1999). This amino acid gets converted to GABA by means of a decarboxylation reaction catalyzed by glutamic acid decarboxylase under conditions oflow pH (Dover and Halpern, 1972a). Glutamic acid decarboxylase deficient mutants are sensitive to acid shock (De Biase et al., 1999). Therefore, it is likely that GABA confers acid resistance to bacteria. To check for the possible role ofGABA 31 in acid resistance, we tested the sensitivity of our GABA utilizing mutant strains (from

Table 2b) along with appropriate controls (from Tables 2a & 2b) to various weak acids.

Sensitivity was tested against sodium acetate (5-50 mM), benzoate (2-10 mM) and salicylate ( 1-5 mM). These weak acids were filter sterilized and the pH was adjusted to

7.0. Minimal salts medium was used with ammonium salts as the nitrogen source and glucose (0.2%) as the carbon source. Proline, thiamine and MgS04 were added as indicated previously. Plates were made with and without GABA (0.2%) to test for

GABA's contribution towards acid adaptation. Cells were replica plated onto plates and the plates were incubated overnight at 3-r'C.

2.13 Osmoprotection test

Glutamate is also postulated to play a protective role against salt stress (De Biase et al., 1999). Since the pathway ofGABA breakdown in E. coli converges with that of glutamate catabolism by means of a decarboxylation reaction catalyzed by glutamic acid decarboxylase (Dover and Halpe~ 1972a), investigating glutamate's role as an osmoprotectant had relevance towards this project. Hence, to test for glutamate's role as an osmoprotectant, we tested the sensitivity of our GABA utilizing mutant strains (from

Table 2b) along with appropriate controls (from Tables 2a & 2b) to varying concentrations of salt (0 mM NaC~ 250 mM NaC~ 500 mM NaCl). Minimal salts medium was used with ammonium salts as the nitrogen source and glucose (0.2%) as the carbon source. Proline, thiamine and MgS04 were added as indicated previously. The test compound (glutamate) was filter sterilized and the pH was adjusted to 7.0 with KOH.

Plates were made with and without glutamate which was added at a concentration of 50 32 rnM. Cells were replica plated onto plates and the plates were incubated overnight at

37°C. 33

Figure 4. Strategy employed to isolate three different groups of GABA utilizing mutant strains directly from the wild type (MC4100) background. 34

MC4100 streaked on LB, incubate 0/N at 37°C l 8 different colonies to inoculate 8 LB broth tubes, let grow 0/N at 37° C to Stationary Phase

Each LB broth tube to streak out a LXS plate, incubate 0/N at 37° C l Independent colonies from each of the 8 LXS plates spread plated onto the 8 different plates for each of the 3 types of selective media, incubate at 37° C for 3-4 days

Spontaneous mutants (big colonies) appearing Group I (CN)-M9 (without NH4CI) + GABA (0. 7°k) Group II (N)-M9 (without NH4CI) + Glucose+ GABA (0.2%) Group Ill (C)-M9 (with NH4CI) + GABA (0.5%)

3 rounds of purification, store cells in glycerol at -80° C 35

Figure 5. Schematic representation of the strategy used to isolate GABA utilizing mutants directly from strains carrying lacZ fusions to members of the gab operon. Mutational frequency (MF) values represent per cell mutational frequency. 36

0 • Spread plating cells from HS1010°, HS1010p , HS1057° and HS1057p0 onto the 3 types of selective media

3 classes of GABA 2 classes of GABA mutants from mutants from HS1057° HS1010° (no N utilizing mutants obtained) ~

CN- MF=2.8x1 a-s per cell CN-- MF=1.2x1Q-6 per cell N - MF=1.3x1Q-5 per cell C - MF=1.6x1 o-s per cell C - MF=2.8x1Q-6 per cell

(No GABA mutants obtained from the rpos- strains) t Selected for over-expressing GABA mutants by streaking 5 mutant colonies of each class onto LXS plates along with appropriate controls CHAPTER3. RESULTS

3.1 GADA utilizing mutants obtained

Many strains of E. coli K-12 cannot utilize GABA as the sole source of carbon or nitrogen (Dover and Halpern, 1972a). Attempts were made to grow MC4100, a wild type strain ofE. coli K-12, in media containing y-aminobutyrate (GABA). This strain tested was unable to grow on most types of selective media supplemented with GABA. The only exception to this rule was found on minimal salts media (without NH4Cl salts) with

GABA (0.2%) and succinate (0.5%). Therefore, it seems that some E. coli K-12 strains, as well as E. coli B strains, can use GABA as a nitrogen source in the presence of a less preferred carbon source, such as succinate, but they cannot degrade GABA as a carbon source (Donnelly and Cooper, 1981).

Dover and Halpern (1972a) were unable to isolate GABA carbon utilizing mutants from wild-type strains. However, using appropriate selective media, we were able to obtain GABA utilizing mutants of all three classes directly from the wild-type strain (MC4100). We were able to isolate mutants which could use GABA as a carbon and nitrogen source or solely as a carbon or nitrogen source alone. The phenotype of these GABA utilizing mutants was confirmed by replica plating the strains on selective media (Table 3).

37 38

3.2 (3-galactosidase assay using transduced GADA mutants

It is documented in the literature that GABA utilizing mutants exhibit a six- to nine-fold increase in the activities of the gab encoded functions compared to wild-type E. coli K-12 strains which exhibit very low activity of these gab genes (Dover and Halpern,

1972a). It is therefore postulated that this over-expression may play an important role for these mutants to be able to utilize GABA. Hence, in order to characterize the GABA utilizing mutants that we isolated and to identify the mutation(s) responsible for over­ expression of the gab operon, these mutants were transduced with lacZ fusions to the two different gab genes by PI-mediated transduction as described in Materials and Methods

(see Table 2a for strain list).

These transductants were then utilized in lac fusion studies. (3-galactosidase assays were performed as descnbed in Materials and Methods on isogenic wild type and rpoS strains along with the three groups of GABA utilizing mutant strains carrying a gabP: :lacZ reporter gene fusion (Figure 6). Expression studies revealed that these transduced GABA utilizing mutant strains were not over-expressing the functions of the gab operon. In filet, expression levels were similar to wild type expression. In rich media, (3-galactosidase activity for rpos+" strains was very low in exponential phase and was induced approximately two to three fold upon entry into stationary phase. In rpoS strains, total activity was greatly reduced in both phases. This supports previous findings by Schellhorn et al. (1998) which argue in favor of growth-phase and RpoS-dependent regulation of the gab genes. 39

Since the addition of GABA has been shown to be an inducer of gab expression in E. coli (Dover and Halpe~ 1972b) on minimal salts media, we also tested whether the expression of the gab operon was specifically induced by this amino acid on rich media

(Figure 6). However, the results indicate that GABA is not an inducer of gab operon expression when cells are growing in LB media

3.3 Sequence analysis of the gab promoter in the GABA mutants

We wanted to explore the reasons for suppression of over-expression of the gab operon from our transduced GABA mutants (Figure 6). To determine if this could be due to the replacement of the mutation site with the wild-type site during the transduction event, we proposed that the mutation may lie somewhere within the linkage distance of the gab operon. Figure 1 shows a detailed schematic of the gab region. The gab operon has been previously identified as a member of the RpoS regulon (Schellhorn et al., 1998).

It lies two ORFs downstream ofthe csiD gene, a carbon starvation-inducible gene in E. coli, that has been shown to be highly RpoS dependent (Marschall et al., 1998) and whose expression is driven from a single upstream promoter. This RpoS regulated promoter is the first promoter upstream of the gab operon (see fig. 1) and is therefore likely to drive expression of the gab genes. As a result, we designed primers as descn'bed in Materials and Methods to amplify this gab promoter in our GABA utilizing mutant strains in order to identifY potential mutation( s) that could be responsible for over­ expression of the gab genes thereby allowing GABA utilization. Wild-type sequence of the csiD promoter region was independently verified in our lab by sequencing this region in MC4100 wild-type strains and was found to be identical to the sequence provided by 40

Marschall et al. ( 1998). Comparison of the amplified products to wild type sequence was done using Clustal W sequence alignment program. The results (Figure 7) show that there was 100 % sequence identity between the wild-type and GABA utilizing mutant strains. Hence, no mutation was found in the csiD promoter region in the GABA utilizing mutants.

3.4 Dispensability of the gab operon for GADA utilization

To detennine if the gab operon is a requirement for GABA transport and metabolism, we attempted to isolate GABA utilizing mutants from wild type fusion strains containing lacZ fusions to the two different gab genes (HS 1010° and HS 105r} and their rpoS derivatives (HS1010p0 and HS1057p~. Hence, ifthe gab genes are necessary for GABA utilization, then we do not expect to obtain any GABA utilizing mutants from strains carrying mutations in the gab genes. Surprisingly, we were able to obtain GABA mutants from both these fusion strains in the rpos*" background, but not from the rpoS background (Table 4). We were able to obtain all three classes ofGABA mutants when gabP was disrupted and two classes of GABA mutants despite the gabD mutation. Interestingly, we did not obtain any nitrogen utilizing GABA mutants from strains carrying a mutation in gabD. This observation directly contradicts earlier findings by Metzer and Halpern (1979) where GABA nitrogen utilizing mutants obtained sustained normal transport and traosamination but were completely devoid of succinic semialdehyde dehydrogenase activity. Furthermore, no GABA utilizing mutant strains were obtained from any of the rpoS fusion strains used or the rpoS non-fusion control strain (HS 1600). Taken together, these results suggest that the gab operon is dispensable 41 for GABA utilization and that other RpoS regulated functions are possibly important in

GABA uptake and metabolism.

Of the GABA mutants obtained, five mutant colonies from each class were tested on LB-X-gal plates to select for those mutants that appeared to be over-expressing the gab genes. Table 5 shows the percentage ofGABA mutants obtained within each class that are potentially over-expressers. Table 2b is a list of these mutant strains that were stocked. For reasons that are not clear, I had difficulty replicating the over-expression that was evident on LB-X-gal plates (Figure 8) in liquid LB media (Figure 9). In fact, the liquid assay results indicate that gab expression levels of the GABA mutants are lower than the wild type fusion strain (HS 1057} which cannot utilize GABA Furthermore, the expression levels ofthese mutants are similar to those of the rpoS:fusion strain

(HS 1057p}, except in stationary phase where activity is about five fold higher for the

GABA utilizing mutants.

Induction Studies

We wanted to test putative inducers of gab operon expression to obtain some regulatory information about this RpoS dependent regulon. Both conditioned media

(Baca-DeLancey et al., 1999) and y-aminobutyrate (Dover and Halpern, 1972b) have been implicated as inducers ofgab expression. 13-galactosidase assays were done in LB media as descn'bed in Materials and Methods with the potential inducing substance added to the growth medium. 42

3.5 Putative inducers of gab expression:

3.5.1 Conditioned Medium

Baca-DeLancey et al. (1999) showed that one of the gab genes, gabT, which codes for y-aminobutyrate transaminase, is induced by conditioned medium at mid­ logarithmic phase. According to that study, the induction ofgab Tin the presence of conditioned medium vs. LB was 19-fold at mid-logarithmic phase, but this induction became less pronounced as cells reached high density. To determine ifthe other two members of this cluster elicited a similar response in the presence of conditioned medium, we used the two wild type fusion strains carrying lacZ fusions to the two remaining gab genes (HS I 0 I 0° and HS I 057} along with their rpoS derivatives

(HS1010p0 and HS1057p} in this induction experiment.

Induction studies revealed that neither gabD nor gabP is induced by conditioned medium (Figure I 0). In fact, induction in conditioned medium happens at the same time as in I X LB. Nevertheless, our results suggest that expression of these gab genes was higher in conditioned medium vs. LB (Table 6). Expression levels for gabP were similar at mid-logarithmic phase, but 2-fold higher in conditioned medium vs. LB at stationary phase, and for gabD, expression levels were 25-fold higher in conditioned medium vs.

LB at mid-exponential phase, but only 2-fold higher in conditioned medium at stationary phase. In rpoS strains, activity was greatly reduced in both types of media. Thus, there seems to be a factor present in conditioned medium which causes gab expression to be higher compared to LB media. Furthermore, this factor seems to exert its effect through

RpoS. 43

3.5.2 LB media supplemented with GABA (0.2%)

Previous studies seem to indicate that y-aminobutyrate may be an inducer of gab expression. For instance, it has been shown that when various strains of E. coli are grown on y-aminobutyrate as the sole nitrogen source, an NAD(P)-dependent succinate semialdehyde dehydrogenase is induced coordinately with 4-aminobutyrate:2- oxogluterate transaminase (Dover and Halpern, 1972b). Furthermore, Donnelly and

Cooper (1981) showed that when mutants are grown with y-aminobutyrate as the sole nitrogen source, an NADP-dependent enzyme is induced. These observations were made when strains were growing in minima1 salts media. To determine ifthe inducing effect of

GABA was detectable in LB media, we used the wild type fusion strain carrying the lacZ fusion to gabD (HS1057j, along with its rpoS derivative (HS1057p~, in this induction experiment. The assays were done in duplicate and the results are plotted in Figure 11.

Our data (Figure 11) suggests that GABA does not induce expression of the gab genes when cells are growing in LB media. Expression studies revealed that induction in media containing GABA (0.2%) happens at the same time as in unsupplemented rich media. Furthermore, unlike the response to conditioned medium, in rich media GABA does not increase gab expression levels (Table 7). As a result, expression levels of the gab gene were similar in LB media in the presence or absence ofGABA. For the rpoS strain, ~-galactosidase activity in the absence ofGABA increased 9-13 fold as cells transitioned from mid-exponential phase to stationary phase. In the presence of GABA, activity increased from 6-11 fold as cells went from mid-exponential phase to stationary phase. In the rpoS strain, activity was greatly reduced in both phases in both types of 44 media These results support our findings using transduced GABA mutants in a previous experiment (Figure 6).

3.5.3 Minimal Media supplemented with GABA (0.2%)

To confirm the inducing effects ofGABA in minimal salts media as suggested by previous findings (Dover and Halpern, 1972b; Donnelly and Cooper, 1981), we replica plated GABA mutants (from Table 2b) along with appropriate controls (MC4100,

0 HS1600, HS1010°, HS1010p , HS105r and HS1057p~ on minimal plates as descnbed in Materials and Methods. The results are shown in Figure 12. The replica plating results fail to show any indications of GABA being an inducer ofgab expression when cells are grown in minimal salts media In fact, minimal plates in the absence of GABA with ammonium as the nitrogen source and either glucose (0.2%) or succinate (0.5%) as the carbon source appear darker than plates supplemented with GABA to a final concentration of 0.2%. These findings are in conflict with earlier work done by other researchers (Dover and Halpern, 1972b; Donnelly and Cooper, 1981).

3.6 Role of glutamate in acid adaptation and osmotoleranee

Glutamate, an amino acid, is converted to GABA by means of a decarboxylation reaction catalyzed by glutamic acid decarboxylase. This enzyme is encoded by the gad genes, gadA and gadB. It has been proposed that expression of these gad genes is positively regulated by acidic shock and salt stress (De Biase et al., 1999). Hence, to confirm the role of glutamate in acid resistance and osmotolerance, we tested the sensitivity of our GABA utilizing mutant strains obtained (from Table 2b) along with 45 appropriate wild type controls (from Tables 2a and 2b) to various weak acids and varying concentrations of salt.

For the acid adaptation test, sensitivity was tested against various weak acids.

These included sodium acetate (5-50mM), benzoate (2-10 mM) and salicylate (1-5 mM).

Sensitivity was tested on minimal plates in the presence and absence of GABA (0.2%) as descnbed in Materials and Methods. The results are summarized in Table 8. For sodium acetate, maximal growth was observed up to 45 mM of acid. However, surprisingly at 50 mM of acid, cells grew better in the absence of GABA Likewise, at 10 mM benzoate, cells exhibited enhanced growth in the absence of GABA A similar phenomenon was seen in the case of 1 mM salicylate where having GABA in the media proved to be detrimental for growth. Interestingly, at this concentration of salicylate, no growth was observed for GABA utilizing mutant strains carrying a gabD-lacZ mutation (HS 1903a,

HS1903b, HS1903c), regardless of whether GABA was present or absent from the media.

Furthermore, none of the strains were able to grow at higher concentrations of salicylate.

For the osmoprotection test, sensitivity was tested against varying concentrations of salt (0-500 mM NaCl) on minimaJ plates in the presence and absence of glutamate (50 mM) as descnbed in Materials and Methods. The results are summarized in Table 9.

Maximal growth was observed for wild type and mutant cells at 0 mM NaCl in the presence or absence of glutamate. However, as suspected, at 250 mM NaCl growth was markedly enhanced in media containing glutamate. Furthermore, at 500 mM NaCl no growth was observed in the absence of glutamate. Interestingly, in the presence of 46 glutamate at this highest concentration of salt, all strains except the GABA utilizing mutants carrying gabD-lacZ fusions were able to grow. 47

Table 2a. E. coli strains used in this study.

Strains Genotype Source/Reference A) Strains

MC4100 A(argF-IacZ)205 araD139 flbB5301 re/A 1 rpsL 150 thi ptsF25 G. Weinstock

HS1600 as MC4100 butrpoS13::Tn10 H. Schellhorn

HS1700 as MC41 00 but selected to use GABA as C and N source R. Kasra

HS1701 as MC41 00 but selected to use GABA as N source R. Kasra

HS1702 as MC41 00 but selected to use GABA as C source R. Kasra

HS1710 as HS1700 but gabP::A.p/acMu53 [ljl{gabP-/acZ)1] H. Schellhorn

HS1711 as HS1701 but gabP::A.p/acMu53 [ljl{gabP-/acZ)1] H.Schellhorn

HS1712 as HS1702 but gabP::A.p/acMu53 [ljl{gabP-/acZ)1] H. Schellhorn

HS1713 as MC4100 but gabP::A.p/acMu53 [ljl(gabP-/acZ)1] H. Schellhorn and rpoS13::Tn10

HS1714 as MC4100 but gabP::A.p/acMu53 [ljl{gabP-/acZ)1] H. Schellhorn

HS1720 as HS1700 but gabD::A.p/acMu53 [ljl(gabD-/acZ)1] H. Schellhorn

HS1721 as HS1701 but gabD::A.p/acMu53 [ljl{gabD-/acZ)1] H. Schellhorn

HS1722 as HS1702 but gabD::A.p/acMu53 [ljl(gabD-/acZ)1] H. Schellhorn

HS1723 as MC4100 but gabD::A.p/acMu53 [ljl(gabD-/acZ)1] and rpoS13::Tn10 H. Schellhorn

HS1724 as MC4100 but gabD::A.p/acMu53 [ljl{gabD-/acZ)1] H. Schellhorn

HS1700P as HS1700 but rpoS13::Tn10 R. Kasra 48

Table 2a (continued). E. coli strains used in this study.

Strains Genotype Source/Reference

HS1701P as HS1701 but rpoS13::Tn10 R. Kasra

HS1702P as HS1702 but rpoS13::Tn10 R. Kasra

HS1710P as HS1710 but rpoS::Tn10 H. Schellhorn

HS1711P as HS1711 but rpoS::Tn1 0 H. Schellhorn

HS1712P as HS1712 but rpoS::Tn10 H. Schellhorn

HS1720P as HS1720 but rpoS::Tn10 H. Schellhorn

HS1721P as HS1721 but rpoS::Tn10 H. Schellhorn

HS1722P as HS1722 but rpoS::Tn10 H. Schellhorn

B) Transducing Phage

P1 vir(HS1010) contains /acZ fusion to gabP promoter laboratory stock

P1 vir (HS1 057) contains /acZ fusion to gabD. promoter laboratory stock 49

Table 2b. E. coli strains used in this study.

Strains Genotype Source/Reference

A) Strains

HS1010° as MC4100 but A.~gabP-IacZ)1 transduced from MC4100 by L.C.Wei

HS1010p0 as HS1010° but rpoS13::Tn10 L.C.Wei

HS105r as MC4100 but ~gabD-IacZ)1 L.C.Wei

HS1057p0 as HS1osr but rpoS13::Tn10 L.C.Wei

HS1900b as HS1010° but selected to use GABA as C and N source this study

HS1900c as HS1010° but selected to use GABA as C and N source this study

HS1900e as HS1010° but selected to use GABA as C and N source this study

HS1900g as HS1010° but selected to use GABA as C and N source this study

HS1901a as HS1010° but selected to use GABA as N source this study

HS1903a as HS1 osr but selected to use GABA as C and N source this study

HS1903b as HS1 osr but selected to use GABA as C and N source this study

HS1903c as HS1 osr but selected to use GABA as C and N source this study

B) Transducing Phage

P1vir(HS1010) contains /acZfusion to gabP laboratory stock

P1 vir (HS1 057) contains lacZ fusion to gabD laboratory stock 50

Table 3. Analysis of GABA mutants obtained for utilization of GABA as a Carbon and/or Nitrogen source. a

Group Selected to use GABA as: Ability to use GABA as: Nsource Csource N c CN (CN) + + + + +

II (N) + +

Ill (C) + + + + Results from replica plating experiment. Poor or no growth(-); good growth(+} 51

Table 4. GABA mutants obtained from wild type fusion backgrounds.a

6 Parental Strain GABA mutants mutant colonies Mutational frequencyc

HS1010° CN 55 2.8 X 10-a

N 200 1.3 X 10-s

c 55 2.8 X 10-a

HS105-,'l CN 23 1.2 X 10-a

N 0 0

c 31 1.6 X 10-a

Results from spread plating experiment as described in Materials and Methods. bNo GABA mutants were obtained from the rpoS" fusion strains. 'Values for per cell mutational frequency were calculated as described in Materials and Methods. 52

Table 5. Over-expressing GABA mutants obtained.a.b

Strain Genotype percent of the total populatlonc

HS1900 as HS1010° but CN GABA mutant 80%

HS1901 as HS1010° but N GABA mutant 20%

HS1903 as HS105~ but CN GABA mutant 60% Over-expressers were selected by comparing lac expression to the appropriate wild type fusion strain on LB-X-gal plates as described in Materials and Methods. bNo over-expressing C utilizing GABA mutants were obtained from HS1010°; also, no over-expressing CorN utilizing GABA mutants were obtained from HS105~. cPercent values for over-expressers are calculated from a total of five colonies of each group. 53

Table 6. Expression of gab genes in response to conditioned medium.

(}-galactosidase accumulation8

- Conditioned Medium + Conditioned Medium

Strain Class Mid-log6 Statlonaryc Mid-log Stationary

HS1010° WT (gabP-IacZ) 30.4 23.3 30.5 67.1

HS1010p0 rpos- (gabP-IacZ) 0.7 0.7 1.2 1.1

HS105~ WT (gabD-/acZ) 3.0 51.1 75.3 149.5

HS1057p0 rpoS (gabD-IacZ) 1.6 1.9 2.6 2.5

Specific activity was determined in Miller Units. bMid-logarithmic phase was determined in cells to be at 00600 of 0.4-0.6. cstationary phase was determined in cells to be at ODsoo of 1.0-1.6. 54

Table 7. Expression of gab genes in response to GABA (0.2%).8

p-galactosidase

-GABA +GABA

Strain Class Mid-loge Stationaryd Mid-log Stationary

HS105fl WT 12.30 135.10 15.05 130.10 (gabD-IacZ)

HS1057p0 rpoS 3.15 4.75 3.6 6.45 (gabD-IacZ)

8 (J-galactosidase assay was done in LB media supplemented with GABA (0.2%) as described in Materials and Methods. bSpecific Activity was expressed in Miller Units. cMid-logarithmic phase was determined in cells 00600 of 0.4-0.6. dStationary phase was determined in cells 00600 of 2.0-2.5. Values represent the average determined from two independent samples. 55

Table 8. Growth of wild-type and GABA mutant strains on various weak acids in the presence and absence of y-aminobutyrate.8

Strain Class Sodium Acetate6·a Benzoate Salicylate +GABA -GABA +GABA -GABA +GABA -GABA 5 45 50 5 45 50 2 6 10 2 6 10 1 3 5 1 3 5 MC4100 WT ++ ++ + ++ ++ ++ ++ ++ + ++ ++ ++ + - - ++ --

HS1600 rpoS" ++ ++ + ++ ++ ++ ++ ++ + ++ ++ ++ + - - ++ --

HS1010° gabP ++ ++ + ++ ++ ++ ++ ++ + ++ ++ ++ + - - ++ --

0 HS1010p rpoS"gabP ++ ++ + ++ ++ ++ ++ ++ + ++ ++ ++ + - - ++ --

HS105fl gabD- ++ ++ + ++ ++ ++ ++ ++ + ++ ++ ++ + - - ++ --

0 HS1057p rpoS-gabD- ++ ++ + ++ ++ ++ ++ ++ + ++ ++ ++ + - - ++ --

HS1900b CNGABA mutantgabP ++ ++ + ++ ++ ++ ++ ++ + ++ ++ ++ + - - ++ --

HS1900c CNGABA mutantgabP ++ ++ + ++ ++ ++ ++ ++ + ++ ++ ++ + - - ++ --

HS1900e CNGABA mutantgabP ++ ++ + ++ ++ ++ ++ ++ + ++ ++ ++ + - - ++ --

HS1900g CNGABA mutantgabP ++ ++ + ++ ++ ++ ++ ++ + ++ ++ ++ + - - ++ --

HS1901a NGABA mutantgabP ++ ++ + ++ ++ ++ ++ ++ + ++ ++ ++ + - - ++ --

HS1903a CNGABA mutant gabD- ++ ++ + ++ ++ ++ ++ ++ + ++ ++ ++

HS1903b CNGABA mutant gabD- ++ ++ + ++ ++ ++ ++ ++ + ++ ++ ++

HS1903c CNGABA mutant gabD- ++ ++ + ++ ++ ++ ++ ++ + ++ ++ ++ Results from replica plating experiment. bno growth(-); weak growth(+); strong growth(++) cConcentrations of acids used are in mM. 56

Table 9. Growth of wild-type and GABA mutant strains on varying concentrations of salt in the presence and absence of glutamate. a

+Glutamate -Glutamate Strain Class OmM 250mM 500mM OmM 250mM 500mM

MC4100 wr ++ ++ + ++ +

HS1600 ~ ++ ++ + ++ +

HS1010° gabP ++ ++ + ++ +

0 HS1010p ~gabP ++ ++ + ++ +

HS105-,o gabU ++ ++ + ++ +

0 HS1057p rpoS"gabD ++ ++ + ++ +

HS1900b CN GABA mutant gabP ++ ++ + ++ +

HS1900c CN GABA mutant g8bP ++ ++ + ++ +

HS1900e CN GABA mutant gabF ++ ++ + ++ +

HS1900g CN GABAmutantgabP ++ ++ + ++ +

HS1901a N GABA mutant gabP ++ ++ + ++ +

HS1903a CN GABA mutant gabD ++ ++ ++ +

HS1903b CN GABAmutantgabO ++ ++ ++ +

HS1903c CN GABA mutant gabD ++ ++ ++ +

1Results from replica plating experiment. bno growth(-); weak growth(+); strong growth(++) 57

Figure 6a. Expression of gabP in wild type strains. Flasks containing either LB broth (control treatment) or LB + GABA (0.2%) were inoculated with exponentially growing cultures as described in Materials and Methods, sampled periodically as indicated, and assayed for growth (00600) and p-galactosidase activity. GABA was tested as a potential inducer of gab expression when cells are growing in LB media. (A) HS1714 (rpoS+, gabP-IacZ) grown in LB, (B) HS1714 (rpoS+, gabP-IacZ) grown in LB + GABA, (C) HS1713 (rpos-, gabP-IacZ) grown in LB, (D) HS1713 (rpoS", gabP-IacZ) grown in LB + GABA. Symbols: o, Growth (00600); •. p-galactosidase activity; 0/N, overnight. 58

(A) (B) 100 10 100 10 -Ul 'Ul' c:: ·a ·-;::::,- 80 - 80 ~ ...... (!) ~ 1 e 1 e -c:: - -c:: 60 '-' 60 '-' ~ 0 ~ 0 (!) 0 (!) 0 Ul \0 {f.) \0 ~ ~ "0 "0 ';3 40 0 ';3 40 0 0 ci 0 ci 0 0.1 0 0.1 ~ «! "';3 20 "i 20 ~ ~

0 0~01 0 0.01 0 2 4 6 0/N 0 2 4 6 0/N Time (h) Time (h) (C) (D) 100 10 100 10 ~ -;;- ·-8- 80 ·-8- 80 ..... ~ 2 s - 1 ·-- 1 s ~ 60 ! 6 60 0 (!) 0 (!) 0 Ul 0 Ul \0 «! 10 «! "0 "0 ';3 ';3 0 0 40 0 0 40 ci 0 ci 0 o~·1 «! 0.1 «! "';3 "';3 00 20 00 20 ~ ~

0 0.01 0 0.01 0 2 4 6 0/N 0 2 4 6 0/N

Time (h) Time (h) 59

Figure 6b. Expression of gabP in CN GABA utilizing mutant strains. Flasks containing either LB broth (control treatment) or LB + GABA (0.2%) were inoculated with exponentially growing cultures as described in Materials and Methods, sampled periodically as indicated, and assayed for growth (00600) and p-galactosidase activity. GABA was tested as a potential inducer of gab expression when cells are growing in LB media. (A) HS1710 (rpoS+, gabP-/acZ, CN GABA utilizing mutant) grown in LB, (B) HS1710 (rpoS+, gabP-IacZ, CN GABA utilizing mutant) grown in LB + GABA, (C) HS171 Op (rpoS", gabP-IacZ, CN GABA utilizing mutant) grown in LB, (C) HS171 Op (rpoS", gabP-IacZ, CN GABA utilizing mutant) grown in LB + GABA. Symbols: o, Growth (00600); •· f3 -galactosidase activity; 0/N, overnight. 60

(A) (B) 100 10 100 10

ell ell .t:!-- .t:: 80 -- 80 ::J= 0= 1... 1... G) 1 -E ~ 1 --E 60 .._= ·-::;g 60 .._..= ~ 0 .._ 0 G) 0 G) 0 -·ell \0 ell \0 ce ce ·as"'0 40 0 ·as"'0 40 0 .....0 0.1 0 0 0.1 0 ~ uce "';j ta 01) 20 on 20 ~ ~ 0 0.01 0 0.01 0 2 4 6 0/N Time (h) (C) (D) 100 10 100 10

ell .t:!-- ( ell ( .E-- ::J= 80 - 80 1... 0 ~ e 1... - 1 --.._.. ~ 1 60 .._..~ ~.._.. 0= 60 G) 0 ~.._.. 0 \0 G) 0 ~ ell \0 "'0 ce ·as 40 0 '0 0 0 ·as 40 0 0.1 0 0 ~ 0.1 "';j -~ eo 20 ta 20 ~ 00 ~ 0 0.01 0 0.01 0 2 4 6 0/N 0 2 4 6 0/N Time (h) Time (h) 61

Figure &c. Expression of gabP in N GABA utilizing mutant strains. Flasks containing either LB broth (control treatment) or LB + GABA (0.2%) were inoculated with exponentially growing cultures as described in Materials and Methods, sampled periodically as indicated, and assayed for growth (00600) and p-galactosidase activity. GABA was tested as a potential inducer of gab expression when cells are growing in LB media. (A) HS1711 (rpoS+, gabP-IacZ, N GABA utilizing mutant strain) grown in LB, (B) HS1711 (rpoS+, gabP-IacZ, N GABA utilizing mutant strain) grown in LB + GABA, (C) HS1711p (rpoS", gabP-IacZ, N GABA utilizing mutant strain) grown in LB, (D) HS1711p (rpoS", gabP-/acZ, N GABA utilizing mutant strain) grown in LB + GABA.

Symbols: o, Growth (00600); •· ~ -galactosidase activity; 0/N, overnight. 62

(A) (B) 100 10 100 10

----...... <'-1 ----...... <'-1 "2 80 "2 80 :::J ;:J 1-o ,...... 1-o 4.> 1 E ~ 1 !:: - s 60 '-' ·- 60 s ~'-' 0 6 0 4.> 0 4.> 0 l:ll \0 l:ll \0 td td "'0 "'0 "til 40 ci "til 40 ci 0 0.1 0 0 0.1 0 0 0 ('d td ~en 20 ~en 20 ~ ~

0 0.01 0 0.01 0 2 4 6 0/N 0 2 4 6 0/N Time (h) Time (h) (C) (D) 100 10 100 10 ,...... ,...... rll .....rll ·a ( "2 ;:J 80 :::> 80 1-o 1-o 4.> e ~ 1 1 -!:: 1'-" 60 ; 60 0 ~ -0 4.> 0 4.> 0 rn \0 rll \0 td td "'0 "'0 ci 'til 40 ci 'til 40 0 0 0 0. 1 0 0 0.1 g td ~ ~ 01) 20 bO 20 ~ ~ 0 0.01 0 0.01 0 2 4 6 0/N 0 2 4 6 0/N

Time (h) Time (h) 63

Figure 6d. Expression of gabP in C GABA utilizing mutant strains. Flasks containing either LB broth (control treatment) or LB + GABA (0.2%) were inoculated with exponentially growing cultures as described in Materials and Methods, sampled periodically as indicated, and assayed for growth (00600) and P-galactosidase activity. GABA was tested as a potential inducer of gab expression when cells are growing in LB media. (A) HS1712 (rpoS+, gabP-IacZ, C GABA utilizing mutant) grown in LB, (B) HS1712 (rpoS+, gabP-/acZ, C GABA utilizing mutant) grown in LB + GABA, (C) HS1712p (rpos-, gabP-IacZ, C GABA utilizing mutant) grown in LB, (D) HS1712p (rpoS', gabP-IacZ, C GABA utilizing mutant) grown in LB + GABA. Symbols: o, Growth (00600); •. p -galactosidase activity; 0/N, overnight. 64

(A) (B) 100 10 100 10 ,...... _ '- '- -. -. (!) e ~ 1 e 1 ...._.s:: - 60 ~ 60 E. ~ 0 '-' 0 '-' 0 (!) 0 (!) \0 ::::.> .... -. '- (!) 1 1 -. es:: ~ :e - 60 '-' 60 s:: - '-' ·- 0 ·-~ 6 0 '-' 0 (!) \0 (!) 0

0 0.01 0 0.01 0 2 4 6 0/N 0 2 4 6 0/N Time (h) Time (h) 65

Figure 7. Clustal W sequence alignment results of PCR amplified csiD promoter region in wild type (MC4100) strain and the three classes of GABA utilizing mutant strains generated from MC4100. HS1700 strains can use GABA as a carbon and nitrogen source, HS1701 strains can use it as a nitrogen source only, and HS1702 strains can use it as a carbon source only. One GABA utilizing mutant from each of the three classes was sequenced. The nucleotide sequence of the csiD promoter lies between nucleotides 160 and 360 in the amplified sequence. Important structural features of the csiD promoter are indicated as follows: highly conserved nucleotides in the CRP binding site are in bold, -10 and -35 positions of the csiD promoter and transcriptional and translational start sites are underlined. See figure 1 for complete nucleotide sequence and important structural features of the csiD promoter. 66

MC4100 -CGAACTA-----TCTCTATTTATAATTTAATGTTATATCTGCCCCGATAAAACGGGGCA 54 HS1700 -CGA-CTA-----TCTCTATTTATAATTTAATGTTATATCTGCCCCGATAAAACGGGGCA 53 HS1701 -TNNNTTCCGACTTCTCTNTTTATAATTTAATGTTATATCTGCCCCGATAAAACGGGGCA 59 HS1702 CCGCANTN-----TCTGTANCGANNNGTTANTGTTATATCTGCCCCGATAAAACGGGGCA 55 * *** * * *** ***************************** MC4100 G-ATAATATGTTTAGTTTACTAACGGTCATTTTGCAGTGAAGCCATTTACTGTTTTTTAT 113 HS1700 G-ATAATATGTTTAGTTTACTAACGGTCATTTTGCAGTGAAGCCATTTACTGTTTTTTAT 112 HS1701 G-ATAATATGTTTAGTTTACTAACGGTCATTTTGCAGTGAAGCCATTTACTGTTTTTTAT 118 HS1702 GGATAATATGTTTAGTTTACTAACGGTCATTTTGCAGTGAAGCCATTTACTGTTTTTTAT 115 * **********************************************************

MC4100 CGACCAGATAATCTGTTCTCTAATGTTAACTCCCCCTAACCTGTTGCTTTAGTTATTCAT 173 HS1700 CGACCAGATAATCTGTTCTCTAATGTTAACTCCCCCTAACCTGTTGCTTTAGTTATTCAT 172 HS1701 CGACCAGATAATCTGTTCTCTAATGTTAACTCCCCCTAACCTGTTGCTTTAGTTATTCAT 178 HS1702 CGACCAGATAATCTGTTCTCTAATGTTAACTCCCCCTAACCTGTTGCTTTAGTTATTCAT 175 ************************************************************

MC4100 TTCCTGTCTCACTTTGCCTTAATACCCTACGTTAAATGTTACTAA~GTTGC~ 233 HS1700 TTCCTGTCTCACTTTGCCTTAATACCCTACGTTAAATGTTACTAATTTGTTGCTTTTGAT 232 HS1701 TTCCTGTCTCACTTTGCCTTAATACCCTACGTTAAATGTTACTAATTTGTTGCTTTTGAT 238 HS1702 TTCCTGTCTCACTTTGCCTTAATACCCTACGTTAAATGTTACTAATTTGTTGCTTTTGAT 235 ************************************************************

MC4100 CACAA~GAAAACAATATGTCGCTTTTGTGCGCATTTTTCAGAAATGTAGATATTTTTA 293 HS1700 CACAATAAGAAAACAATATGTCGCTTTTGTGCGCATTTTTCAGAAATGTAGATATTTTTA 292 HS1701 CACAATAAGAAAACAATATGTCGCTTTTGTGCGCATTTTTCAGAAATGTAGATATTTTTA 298 HS1702 CACAATAAGAAAACAATATGTCGCTTTTGTGCGCATTTTTCAGAAATGTAGATATTTTTA 295 ************************************************************ MC4100 GATTATGGCTACGAAATGAGCATCGCCATGTCACCCTACATCTCATAAGAGGATCGCTTC 353 HS1700 GATTATGGCTACGAAATGAGCATCGCCATGTCACCCTACATCTCATAAGAGGATCGCTTC 352 HS1701 GATTATGGCTACGAAATGAGCATCGCCATGTCACCCTACATCTCATAAGAGGATCGCTTC 358 HS1702 GATTATGGCTACGAAATGAGCATCGCCATGTCACCCTACATCTCATAAGAGGATCGCTTC 355 ************************************************************

MC4100 TGATGAATGCACTGACCGCCGTACAAAATAACGCTGTCGATTCAGGCCAGGACTATAGCG 413 HS1700 TGATGAATGCACTGACCGCCGTACAAAATAACGCTGTCGATTCAGGCCAGGACTATAGCG 412 HS1701 TGATGAATGCACTGACCGCCGTACAAAATAACGCTGTCGATTCAGGCCAGGACTATAGCG 418 HS1702 TGATGAATGCACTGACCGCCGTACAAAATAACGCTGTCGATTCAGGCCAGGACTATAGCG 415 ************************************************************ MC4100 GATTCACCCTCACCCCGTCGGCGCAATCCCCGCGTCTGCTGGAACTCACCTTCACCGAAC 473 HS1700 GATTCACCCTCACCCCGTCGGCGCAATCCCCGCGTCTGCTGGAACTCACCTTCACCGAAC 472 HS1701 GATTCACCCTCACCCCGTCGGCGCAATCCCCGCGTCTGCTGGAACTCACCTTCACCGAAC 478 HS1702 GATTCACCCTCACCCCGTCGGCGCAATCCCCGCGTCTGCTGGAACTCACCTTCACCGAAC 475 ************************************************************

MC4100 AGACGACCAAACAGTTTCTTGAGCAGGTTGCCGAGTGGCCCGTGCAGGCGCTGGAGTACA 533 HS1700 AGACGACCAAACAGTTTCTTGAGCAGGTTGCCGAGTGGCCCGTGCAGGCGCTGGAGTACA 532 HS1701 AGACGACCAAACAGTTTCTTGAGCAGGTTGCCGAGTGGCCCGTGCAGGCGCTGGAGTACA 538 HS1702 AGACGACCAAACAGTTTCTTGAGCAGGTTGCCGAGTGGCCCGTGCAGGCGCTGGAGTACA 535 ************************************************************

MC4100 AATCGTTTCTGCGTTT----TTGG------553 HS1700 AATCGTTTCTGCGTTT----TCGGANNN 588 HS1701 AATCGTTTCTGCGTTTCGGANNNNNNNNNNNNNNNNNNNNNNNNNNNN 598 HS1702 AATCGTTTCTGCGTTTCGGANNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNN 595 **************** 67

Figure 8. Replica plate pattern of wild type strain HS1057° (rpoS+, gabD­ 0 /acZ), its rpos· derivative HS1057p , and Group I (CN) GABA utilizing mutants generated from the wild type rpoS+ fusion background only (HS1903a, HS1903b, HS1903c). All cells were grown in flasks containing LB broth to an 00600 - 1.0. The cells were then pelleted by centrifugation, washed in 1 X M9 salts (without NH4CI) and replica plated onto a LB-X-Gal plate. The plate was incubated overnight at 37°C. 68

HS1057°

HS1057p0

HS1903a

HS1903b

HS1903c 69

Figure 9. Expression of gabD in wild type and "over-expressing" GABA utilizing mutant strains. This assay was done to duplicate the over-expression evident on LB-X-Gal plates (fig. 8) in liquid LB media. Flasks containing LB broth were inoculated with exponentially growing cultures as described in Materials and Methods, sampled periodically as indicated, and assayed for growth (00600) and p-galactosidase activity. (A) HS1 057° (rpoS+, gabD-/acZ), (B) HS1057p0 (rpos-, gabD-IacZ), (C) HS1903a (rpoS+, gabD-/acZ, CN GABA utilizing mutant), (D) HS1903c (rpoS+, gabD-IacZ, CN GABA utilizing mutant). Symbols: o, Growth (00600); •. p-galactosidase activity; 0/N, overnight. 70 (A) (B)

200 10 200 10 -.. -.. r.f.l r.f.l ·a..... ·a..... ;:J 150 -.. ;:J 150 -.. E E '- 1 c:: '- 1 c:: ~ '-' ~ '-' 0 0 ·- 0 ·- 0 6 \0 6 100 \0

Figure 10. Conditioned medium as an inducer of gab expression. Flasks containing either LB broth (control treatment) or conditioned medium were inoculated with exponentially growing cultures as described in Materials and Methods, sampled periodically as indicated, and assayed for growth (00600) and J3-galactosidase activity. (A) HS1010° (rpoS+, gabP-/acZ) grown in LB, (B) HS1 010° (rpoS+,gabP-IacZ) grown in conditioned medium, (C) HS1 01 Op0 (rpos-, gabP-IacZ) grown in LB, (D) HS1010p0 (rpoS-, gabP-IacZ) grown in conditioned medium, (E) HS1 057° (rpoS+, gabD-IacZ) grown in LB, (F) HS1 057° (rpoS+, gabD-IacZ) grown in conditioned medium, (G) HS1 057p0 (rpoS, gabD-IacZ) grown in LB, (H) HS1 057p0 (rpoS, gabD-IacZ) grown in conditioned medium. Symbols: o, Growth (00600); •. J3-galactosidase activity; 0/N, overnight. 72

(A) (B)

200 10 200 10 .., .., -.-::: 175 - 175 s::: ·--s::: ~ ~ 1-o 150 1-o 150 ~ 1 e ~ 1. e - 125 -s::: 125 -s::: '-" '-" ~ 0 ~ 0 ..,(1.) 100 0 ..,(1.) 0 m \0 m 100 \0 "'0 "'0 "r;j 0 ·;; 0 75 0 75 A 0 0.1 0 0 0.1 0 m 50 m ~ ~ 50 01) 01) ~ 25 ~ 25 0 0 0 2 4 6 Time (h) (C) (D) 200 10 200 10 .., '00' - 175 ·s- 175 ·-- ::::> ~ :5 150 1-o 150 ro ~ ~ 1 125 es::: - 125 es::: - '-" ~ '-" ~ 0 (1.) 0 (1.) 100 0 .., 100 0 .., \0 m \0 m "'0 "'0 "r;j "r;j 75 0 0 75 0 0 0.1 0 0.1 0 0 -~ m 50 ~ 50 ~ 01) 01) ~ ~ 25 25 0 0 0 2 4 6 0 2 4 6 Time (h) Time (h) 73 (E) (F)

200 10 200 10 -. - :::>- 150 150 '- '- -. ~ 1 -.. ~ - 125 s 125 sc::: ~ c::: '-" '-" '-" ~'-" Q) 0 Q) 0

0 O.Q1 0 0.01 0 2 4 6 0/N 0 2 4 6 0/N Time (h) Time (h)

(G) (H) 200 10 200 10 -.. :::> 150 '- 150 ~ -.. -. ~ 1 ~ s 125 ! ·- 125 s ~ 0 6 0 v 0 v 0

0 0.01 0 0.01 0 2 4 6 0/N 0 2 4 6 0/N Time (h) Time (h) 74

Figure 11. GABA as an inducer of gab expression in LB media. Flasks containing either LB broth (control treatment) or LB broth+ GABA (0.2%) were inoculated with exponentially growing cultures as described in Materials and Methods, sampled periodically as indicated, and assayed for growth (00600) and (3-galactosidase activity. Also, the assays were done in duplicate. (A) HS1 057° (rpoS+, gabD-IacZ) grown in LB, (B) HS1 057° (rpoS+, gabD-IacZ) grown in LB, (C) HS1 05~ (rpoS+, gabD-/acZ) grown in LB + GABA, (D) HS1 057° (rpoS+, gabD-/acZ) grown in LB + GABA, (E) HS1057p0 (rpoS", gabD-IacZ) grown in LB, (F) HS1 057p0 (rpoS", gabD-/acZ) grown in LB, (G) HS1057p0 (rpos-, gabD-IacZ) grown in LB + GABA, (H) HS1057p0 (rpos-, gabD-IacZ) grown in LB + GABA. Symbols: o, Growth (00600); •, P-galactosidase activity; 0/N, overnight. P-galactosidase (Nliller Units) ...... P-galactosidase (Miller Units) ,....-..., N ~ ~ N (] ()I 0 ()I 0 ()I ...... > 0 0 - ..._.., 0 0 -0 0 0 0 ~ 0 0 0

N t-3 ..., §' ~ 3-·~ (\) (1) -::r O'l \_ -::r O'l. 0 - ~ ~ 0 0 ...... 0 !=> ...... 0 0 ..... 0 0 :...... I - O.D. 600 (nm) O.D. 600 (nm)

,....-.... p-galactosidase (Miller Units) ...... P-galactosidase (Miller Units) ~ ..... N ~ ..... N cc ()I 0 ()I g ..._.., ()I 0 ()I 0 ..._.., 0 0 0 0 - 0 0 0 0 0 0 0

N N t-3 §' ..., ~ .... "'" (1) 3 (1) ::r - O'l e- Ol ;r b ~T rr ~ 0 p ..... 0 !P ...... 0 0 0 - 0 ~ ...... -.....) Vl O.D. 600 (nm) O.D. 600 (nm) P-galactosidase (M:iller Units) ...-.... P-galactosidase (Miller Units) ...... N ...... N ~ U'l ~ U'l 0 C1l 0 0 0 8 g: 8 "-oo 0 0 0 0 0 "'-"" 0 0

N N ~ ~ §' ~ -·a ~ C1) C1) .- .- 0) 0') '-"=- =- . o-cr=- ---c -o-r b z 0 !=I ...... 0 p ...... 0 ...... 0 .....0 ..... 0 ...... O.D. 600 (nrn) O.D. 600 (nm)

P-galactosidase (M:iller Units) ...-.. P-galactosidase (Miller Units) ..-...... N ...... N ~ (/1 ....._., C1l 0 C1l 0 0 0 8 ~ 8 0 0 0 0 0 "'-" 0 = 0

N 1\) ~ ~ ~ §' ~ §' C1) C1) .- .- en ::l"O) '-" ' '-"=-

~ ..... ~ 0 0 ...... 0 p ~ ...... 0 :..... 0 0 ...... 0 ...... -.l O.D. 600 (nm) 0\ O.D. 600 (nm) 77

Figure 12. GABA as an inducer of gab expression in minimal medium. Flasks containing LB broth were inoculated with overnight cultures as described in Materials and Methods. Cells were grown to early stationary phase (00600 - 1.0), pelleted by centrifugation, washed in 1 X M9 salts (without NH4CI) and replica plated onto various minimal plates as described in Materials and Methods. The test compound (GABA) was added at a final concentration of 0.2%. Replica plating was done in triplicate using the following strains in the following order: MC4100 (WT), HS1600 (rpoS), HS1010° (rpos•, gabP-IacZ), HS1010p0 (rpoS, gabP-IacZ), HS105t»(rpos•, gabD-IacZ), HS1057p0 (rpos-, gabD-/acZ), HS1900b-g (rpos•, gabP-IacZ, CN GABA utilizing mutants), HS1901a (rpos•, gabP-IacZ, N GABA utilizing mutant), HS1903a-c (rpos•, gabD­ IacZ, CN GABA utilizing mutants). 78

Replica Plate Pattern Order of Strains Used : M9+NH4+ MC4100, HS1600 M9+NH4+GLU GLU+GABA HS1010°, HS1010p0 HS1057°, HS1057p0 HS1900b, HS1900c HS1900e, HS1900g HS1901a HS1903a, HS1903b HS1903c

LB LB+GABA

• M9+NH4+SUC M9+N H4+ SUC+GABA

M9+SUC+GABA

M9 M9+GABA CHAPTER 4. DISCUSSION

About 25 different compounds, consisting mainly of amino acids but also including purines, pyrimidines, and nitrate, can serve as sole nitrogen sources for E. coli

(Tyler, 1978) and Salmonella typhimurium (Gutnick et al., 1969). Most ofthe 20 natural

L-amino acids and several D-amino acids serve as potential carbon and energy sources for members of the Enterobacteriaceae. y-aminobutyrate (GABA), a four carbon amino acid, serves as an intermediate in the pathways ofL-glutamate degradation in E. coli

(Marcus et al., 1969) and L-arginine and putrescine catabolism in E. coli and K aerogenes (Friedrich and Magasanik, 1979; Shru"be et al., 1985). GABA has been shown to serve as the sole carbon and nitrogen source in K aerogenes (Friedrich and

Magasanik, 1979) and in E. coli (Dover and Halpern, 1972b). The GABA degradation pathway is similar inK aerogenes (Friedrich and Magasanik, 1979) and in E. coli (Dover et al., 1972b) in that it involves the gab operon, a highly RpoS-dependent regulon

(Schellhorn et al., 1998). However, the gab system has been examined in much more detail in E. coli.

It has been well established that wild type E. coli cannot degrade GABA as the sole carbon source (Dover and Halpern, 1972a) and can utilize it as a nitrogen source only in the presence of succinate (Donnelly and Cooper, 1981). This inability ofwild type strains to grow in the presence of GABA may be dependent on the expression levels

79 80 of the gab genes in these strains. In fact, the enzymes involved in GABA catabolism are expressed at low constitutive levels in wild type strains compared to GABA mutants which exhibit eightfold higher levels of these enzymes (Dover and Halpern, 1972a).

Therefore, the reduced level of gab expression in wild type strains is postulated to be insufficient for the utilization of GABA as sole carbon or nitrogen source in these strains

(Dover and Halpern, 1972a). This seems interesting as it forces one to consider the possible reasons for the presence of the gab operon in these wild type E. coli strains, something which is not clear at this point.

Even though wild type E. coli K-12 strains cannot utilize GABA, it is possible to isolate GABA utilizing mutant strains which can use GABA as sole carbon or nitrogen source. Dover and Halpern (1972a) were able to obtain GABA nitrogen utilizing strains but all attempts to obtain GABA carbon utilizing mutants directly from the wild type strains failed. Subsequently, secondary mutants derived :from these GABA nitrogen utilizing strains were obtained which could use GABA as the sole source of carbon.

Interestingly, unlike Dover and Halpern (1972a), we were able to isolate all three classes of GABA utilizing mutant strains directly from the wild type background (MC41 00) by selection on appropriate minimal media supplemented with GABA (Table 2a). We were able to isolate mutants which could use GABA as a carbon and nitrogen source or solely as a carbon or nitrogen source alone. The reason for why we didn't encounter any difficulty in obtaining GABA carbon utilizing mutants directly :from the wild type strain, as did Dover and Halpern (1972a), may be due to differences in wild type strains used between our study and theirs and differences in strategies used for isolating GABA 81 mutants. Their isolation criteria relied upon the use of ultraviolet mutagenesis (Dover and Halpern, 1972a), whereas we used selective minimal media for isolating GABA utilizing mutant strains.

To confirm the phenotype of our GABA utilizing mutants, these strains were replica plated on selective minimal media supplemented with GABA. The results (Table

3) show some interesting observations. We noticed that mutants selected to use GABA

as the sole carbon source could also use GABA as a nitrogen source, but mutants selected to use GABA as the sole nitrogen source could not use GABA as a carbon source. Based

on this data, we can make two predictions. First, it seems that GABA carbon and nitrogen utilizing strains are phenotypically similar to GABA carbon utilizing mutants.

Secondly, GABA nitrogen utilizing mutants most likely outnumber other groups of

GABA mutant strains.

Next, these GABA mutants obtained were transduced with lacZ fusions to members of the gab operon in order to expedite characterization of the factors

responsible for GABA utilization. We found that the transductants lost the capacity to

utilize GABA The only transduced mutants that still retained the ability to utilize

GABA could only use it as a nitrogen source. This observation strongly suggested that a gab operon dependent GABA utilization pathway was was altered in the mutants. Given

that the gab pathway relies on over-expression of the gab operon for GABA metabolism

(Dover and Halpern, 1972a), we assayed our transductants for f3-galactosidase activity

(Figure 6) to measure gab activity in order to characterize the mutations accomodating

this over-expression sufficient for GABA uptake and catabolism. Our working model 82 was that ifthe transduced mutants were over-expressing the gab genes, then the mutation lay somewhere outside the linkage distance of the transducing phage. Conversely, ifthe transductants showed expression levels similar to wild type expression, then that would suggest one of two possibilities. One, the suppression ofgab over-expression could be a result of the replacement of the mutation site with the wild type site during the transduction event, suggesting that the mutation lay somewhere within the linkage distance of the gab operon. Finally, the second possibility pointed towards some other alternative pathway of GABA utilization in these mutants. The results of f3-galactosidase studies (Figure 6) clearly indicated that gab expression levels of the transduced GABA mutants were similar to wild type expression. In an attempt to uncover potential mutations in close vicinity to the gab operon, we sequenced the csiD promoter in wild type (MC41 00) and the three classes ofGABA mutant strains (HS 1700, HS 1701,

HS 1702) (Figure 7). Both csiD (Marschall et al., 1998) and the gab operon (Schellhorn et al., 1998) have been identified as members ofthe RpoS regulon and are transcn"bed in the same direction (Figure 1). As a result, these two adjacent and highly RpoS dependent transcriptional units may be driven from a single upstream RpoS dependent promoter classified as the csiD promoter, or the gab promoter. However, we didn't find a mutation in the csiD promoter region (Figure 7) suggesting that either the mutation responsible for gab over-expression lies in close proximity to the gab operon outside of the csiD promoter region or that some other pathway of GABA utilization is required in these mutants. Future work should include Northern blot analysis ofgab expression in these wild type and GABA mutant strains to determine which factor is most likely responsible. 83

The literature pertaining to GABA comments on the role of the gab operon in

GABA utilization (Dover and Halpern, 1972a, 1972b, 1974). GABA utilizing mutants exhibit a severalfold increase in gab-encoded functions which is postulated to be sufficient to allow for GABA utilization in these strains (Dover and Halpern, 1972a).

Interestingly, this study speculates the presence of an alternate GABA utilization pathway that may be independent ofgab expression. Hence, to validate the possibility that a second mechanism of GABA utilization exists that doesn't involve the gab operon, it was important to show that the gab operon was dispensable (ie., non-essential) for GABA utilization. To accomplish this purpose, we attempted to isolate GABA mutants from wild type fusion strains (Figure 5) carrying lacZ fusions to members of the gab operon

(HS1010° and HS105r} and their rpoSderivatives (HS1010p0 and HS1057p~. The fact that we were able to obtain GABA utilizing mutants from both fusion strains in the rp~ background only {Table 4), suggests two additional features ofGABA metabolism in E. coli. One, that the gab operon indeed is dispensable for GABA utilization. In addition, other RpoS regulated functions, independent ofgab-encoded functions, may allow the cell to metabolize GABA These may include the ast operon, a five gene operon, involved in degradation of arginine and ornithine to generate succinate and glutamate and the gad genes involved in decarboxylation of glutamate to GABA by means of glutamic acid decarboxylase (Baca-DeLancey et al., 1999; DeBiase et al., 1999). Surprisingly, from amongst these GABA mutants that seemed to be making use of some other GABA utilization pathway, we were able to identifY those mutants that were over-expressing the functions of the gab operon . Over-expressers were identified by comparing lac 84 expression to the appropriate wild type fusion strain on LB-X-gal plates as descnbed in

Materials and Methods. Table 5 lists the number of over-expressing GABA mutants obtained. Figure 8 is an illustration of this over-expression. This raises an interesting question because if the gab operon is not being utilized for GABA degradation, then why is it being over-expressed in these GABA utilizing mutant strains? Perhaps, the gab operon serves another purpose, besides GABA utilization, in some strains of

Enterobacteriaceae. Furthermore, in an attempt to elucidate the mutation responsible for this gab over-expression in these bacterial strains, we sequenced the csiD promoter in these GABA mutants also (data not shown). No mutation was found in this region.

Future work entailing primer extension studies might prove useful in pinpointing the exact location of the mutation responsible for gab over-expression.

The next portion of my thesis involved testing putative inducers of gab expression to obtain some regulatory information about this operon. Conditioned media and y­ aminobutyrate were tested for their inducing effects. Conditioned media has been shown to an inducer of gabT expression (Baca-DeLancey et al., 1999) but failed to elicit a similar response for the other two structural genes in this operon (our study, Figure 10).

This conflict in data may be a result of differences in strains used between our study and those of other researchers (Baca-DeLancey et al., 1999). Likewise, synthesis of the two y-aminobutyrate -degrading enzymes has been shown to be induced by y-aminobutyrate itself when cells are growing in minimal salts media (Dover and Halpern, 1972b;

Donnelly and Cooper, 1981). However, we did not find GABA to be an inducer ofgab operon expression, regardless of whether the cells were grown in minimal salts media 85

(Figure 12) or rich media (Figure 11). Once again, factors such as differences in strains used could be an issue responsible for this conflict in findings observed between our study and others. However, other reasons, which are not clear at this point, may also be at work.

Finally, we wanted to investigate the role of glutamate in acid adaptation and osmotolerance. Glutamate has been documented to serve a protective function against

acidic shock and salt stress (DeBiase et al., 1999). To confirm this effect, we tested the sensitivity of our GABA utilizing mutants to various weak acids and varying

concentrations of salt on minimal plates as descn"bed in Materials and Methods. For acid adaptation, sensitivity was tested in the presence and absence of GABA This is because the pathway of GABA breakdown in E. coli converges with that of glutamate catabolism

by means of a decarboxylation reaction catalyzed by glutamic acid decarboxylase (Dover

and Halpern, 1972a). Glutamic acid decarboxylase deficient mutants are sensitive to acid shock (De Biase et al., 1999). Therefore, our hypothesis was that perhaps the ability to

generate GABA or have access to it confers acid resistance to bacteria. As a result, we expected adding GABA to the media to be beneficial for cell growth. Interestingly though, GABA proved to be detrimental for growth because we found that the cells exhibited enhanced growth in the absence of GABA {Table 8). An explanation for this observation is not clear at this point. As for the role of glutamate in osmotolerance,

sensitivity was tested in the presence and absence of glutamate. We hypothesized

optimal growth in the presence of glutamate and indeed this is what we found (Table 9). 86

Based on our findings, therefore, we can conclude that glutamate serves as an osmoprotectant, but its role in acid tolerance is not clear.

Acknowledgements

I would like to thank Dr. Herb Schellhorn for doing the transductions and Dr. Rouha

Kasra for providing GABA mutant strains. This work was funded from an operating grant to H.E.S. from the Natural Sciences and Engineering Council (NSERC) of Canada. 87

4.1 REFERENCES

Andre, B. and Jauniaux, J.C. (1990). Nucleotide sequence of the yeast UGA1 gene encoding GABA transaminase. Nucleic Acids Res. 18,3049.

Awapara, J., Landua, AJ., Fuerst, R, and Seale, B. (1950). Free y-aminobutyric acid in brain. J.BioLChem. 187,35-39.

Baca-DeLancey, R, South, M.T., Ding, X., and Rather, P. N. (1999). Escherichia coli genes regulated by cell-to-cell signaling. Proc. Natl Acad. Sci 96,4610-4614.

Balke, V.L. and Gralla, J.D. (1987). Changes in the linking number ofsupercoiled DNA accompany growth transitions in Escherichia coli. J.Bacteriol 169,4499-4506.

Barth, M., Marschall, C., Muftler, A., Fischer, D., and Hengge-Aronis, R (1995). Role for the histone-like protein H-NS in growth phase-dependent and osmotic regulation of sigma S and many sigma S-dependent genes in Escherichia coli. J.Bacteriol. 177, 3455-3464.

Bartsch, K., Johnn-Marteville, A, and Schulz, A (1990b). Molecular analysis of two genes of the Escherichia coli gab cluster: nucleotide sequence ofthe succinic glutamate:succinic semialdehyde transaminase gene (gab1) and characterization ofthe semialdehyde dehydrogenase gene (gabD). J.Bacteriol. 172, 7035-7042.

Becker-Hapak, M., Troxtel, E., Hoerter, J., and Eisenstark, A (1997). RpoS dependent overexpression of carotenoids from Erwinia herbicola in OxyR deficient Escherichia coli. Biochemical & Biophysical Research Communications.

Betteridge, P.R, and Ayling, P.D. (1976). The regulation of glutamine transport and glutamine synthetase in Salmonella typhimurium. J.Gen.Microbiol 95, 324-334.

Bouvier, J., Gordia, S., Kampmann, G., Lange, R, Hengge-Aronis, R, and Gutierrez, C. (1998). Interplay between global regulators ofEscherichia coli: effect ofRpoS, Lrp and H-NS on transcription ofthe gene osmC. Molec.Microbiol 28, 971-980.

Brechtel, C.E., Hu, L., and King, S.C. (1996). Substrate specificity of the Escherichia coli 4-aminobutyrate carrier encoded by gabP. J.Biol.Chem. 271, 783-788.

Cunin, R, Glansdorff, N., Pierard, A., and Stalon, V. (1986). Biosynthesis and metabolism of arginine in bacteria. Microbiol.Rev. 50, 314-352. 88

DeBiase, D., Tramont~ A., Bossa, F., and Visca; P. (1999). The response of stationary­ phase stress conditions in Escherichia coli: role and regulation of the glutamic acid decarboxylase system. Molec.Microbiol. 6, 1198-1211.

Donnelly, M.I. and Cooper, RA (1981). Succinic semialdehyde dehydrogenases of Escherichia coli. Eur.J.Biochem. 113, 555-561.

Dover, S. and Halpern, Y.S. (1972a). Utilization ofy-aminobutyric acid as the sole carbon and nitrogen source by Escherichia coli K-12 mutants. J.Bacteriol. 109, 835-843.

Dover, S. and Halpern, Y.S. (1972b). Control ofthe pathway ofGABA breakdown in Escherichia coli K-12. J.Bacteriol. 110, 165-170.

Dover, S. and Halpern, Y.S. (1974). Genetic analysis of the gamma-aminobutyrate utilization pathway in Escherichia coli K.-12. J.Bacteriol. 117, 494-501.

Farres, J., Guan, K.L., and Weiner, H. (1989). Primary structures of rat bovine liver mitochondrial aldehyde dehydrogenases deduced from eDNA sequences. Eur.J.Biochem. 180, 61-14.

Friedrich, B., and Magasanik, B. (1978). Utilization of arginine by Klebsiella aerogenes. J.Bacteriol. 133, 680-685.

Friedrich, B., and Magasanik, B. (1979). Enzymes of agmatine degradation and the control oftheir synthesis in Klebsiella aerogenes. J.Bacteriol. 137, 1127-1133.

Fujita, M., Tanaka, K., Takahas~ H., and Amemur3; A (1994). Transcription of the principal sigma-fitctor genes, rpoD and rpoS, in Pseudomonas aeruginosa is controlled according to the growth phase. Molec.Microbiol. 13, 1071-1077.

Gilson, E., Clement, J.M., Brutlag, D., and Hofuung, M. (1984). A family of dispersed repetitive extragenic palindromic DNA sequences in E. coli. EMBO J. 3, 1417- 1421.

Gutnick, D., Calvo, J.M., Klopotowski, T., and Ames, B.N. (1969). Compounds which serve as the sole source of carbon or nitrogen for Salmonella typhimurium LT -2. J.Bacteriol. 100, 215-219.

Guyton, AC., Hall, J.E. (1996). Nervous System: General Principles and Sensory 89

Physiology. Textbook ofMedical Physiology, 565-580.

Hengge-Aronis, R (1996b). Regulation of gene expression during entry into stationary phase. In Escherichia coli and Salmonella: Cellular and Molecular Biology. F.C. Neidhardt, ed. (Washington, D.C.: ASM Press), pp. 1497-1512.

Hengge-Aronis, R, and Fischer, D. (1992). Identification and molecular analysis of gigS, a novel growth phase-regulated and rpoS-dependent gene involved in glycogen synthesis in Escherichia coli. Molec.MicrobioL 6, 1877-1866.

Hengge-Aronis, R, Klein, W., Lange, R, Rimmele, M, and Boos, W. (1991). Trehalose synthesis genes are controlled by the putative sigma factor encoded by rpoS and are involved in stationary phase thermotolerance in Escherichia coli. J.BacterioL 173, 7918-7924.

Hoffinann, W. (1985). Molecular characterization ofthe CAN1locus in Saccharomyces cerevisiae. A transmembrane protein without N-terminal hydrophobic signal sequence. J.Biol.Chem. 260, 11831-11837.

Honore, N., and Cole, S.T. (1990). Nucleotide sequence of the aroP gene encoding the general aromatic amino acid transport protein ofEscherichia coli K-12: homology with yeast transport proteins. Nucleic Acids Res. 18, 653.

Hsu, L.C., Bendel, RE., and Yoshida, A (1988). Genomic structure of the human mitochondrial aldehyde dehydrogenase gene. Genomics 2, 57-65.

Iriarte, M., Stainier, 1., and Comelis, G.R (1995). The rpoS gene from Yersinia enterocolitica and its influence on expression of virulence factors. Infection & Immunity. 63, 1840-1847.

Jacob, F., and Monod, J. (1961). On the regulation of gene activity. Cold Spring Harbor Symp. Quant. Bioi. 26, 193-209.

Jakoby, W.B., and Scott, E.M. (1959). Aldehyde oxidation. III. Succinic semialdehyde dehydrogenase. J.Biol.Chem. 234, 937-940.

Jishage, M., Iwata, A, Ueda, S., and lshihama, A (1996). Regulation of RNA polymerase sigma subunit synthesis in Escherichia coli: intracellular levels of four species of sigma subunit under various growth conditions. J.Bacteriol. 178, 5447-5451. 90

Kaasen, 1., Falkensberg, P., Styrvold, O.B., and Strom, AR. (1992). Molecular cloning and physical mapping of the otsBA genes, which encode the osmoregulatory trehalose pathway of Escherichia coli: evidence that transcription is activated by katF. J.Bacteriol 174, 889-898.

Kahane, S., Levitz, R., and Halpern, Y.S. (1978). Specificity and regulation ofGABA transport in Escherichia coli. J.Bacteriol. 135, 295-299.

Kalman, M., Gentry, D.R., and Cashel, M. (1991). Characterization ofthe Escherichia coli K-12 gltS glutamate permease gene. Mol.Gen.Genet. 225,379-386.

King, S.C., Fleming, S.R., and Brechtel, C. (1995b). Pyridine carboxylic acids as inhibitors and substrates of the Escherichia coli gab permease encoded by gabP. J.Bacteriol. 177, 5381-5382.

King, S.C., Fleming, S.R., and Brechtel, C. (1995a). Ligand recognition properties of the Escherichia coli 4-aminobutyrate transporter encoded by gabP. Specificity of Gab permease for heterocyclic inhibitors. J.BiolChem. 270, 19893-19897.

Kowarz, L., Coynault, C., Robbe-Saule~ V., and Norel, F. (1994). The Salmonella typhimurium katF (rpoS) gene: cloning, nucleotide sequence, and regulation of spvR and spvABCD virulence plasmid genes. J.Bacteriol. 176, 6852-6860.

Kustu, S.G., McFarland, N.C., Hiu, S.P., Esmon, B., and Ames, G.F.-L. (1979). ·Nitrogen control in Salmonella typhimurium: co-regulation of synthesis of glutamine synthetase and amino acid transport systems. J.Bacteriol. 138, 218-234.

Lange, R. and Hengge-Aronis, R. (199Ia). Growth phase regulated expression of bolA and morphology of stationary-phase Escherichia coli cells are controlled by the novel sigma mctor sigmaS. J.Bacteriol 173,4474-4481.

Lange, R. and Hengge-Aronis, R. (1991b). Identification of a central regulator of stationary-phase gene expression in Escherichia coli. Molec.Microbiol 5, 49-59.

Lange, R. and Hengge-Aronis, R. (1994b). The cellular concentration of the sigmaS subunit of RNA polymerase in Escherichia coli is controlled at the levels of transcription, translation, and protein stability. Genes.Develop. 8, 1600-1612.

Lange, R., Barth, M., and Hengge-Aronis, R. (1993). Complex transcriptional control of the sigmas-dependent stationary-phase-induced and osmotically regulated osmY (csi-5) gene suggests novel roles for Lrp, cyclic AMP (cAMP) receptor protein- 91

cAMP complex, and integration host factor in the stationary phase response of Escherichia coli. J.Bacteriol. 175,7910-7917.

Lee, I.S., Lin, J., Hall, H.K., Bearson, B., and Foster, J.W. (1995). The stationary-phase sigma factor sigma S (RpoS) is required for a sustained acid tolerance response in virulent Salmonella typhimurium. Molec.Microbiol. 17, 155-167.

Magasanik, B. (1961). Catabolite repression. Cold Spring Harbor Symp. Quant. Bioi. 26, 249-256.

Manna, D., and Gowrishankar, J. (1994). Evidence for involvement of proteins HU and RpoS in transcription ofthe osmoresponsive proU operon in Escherichia coli. J.Bacteriol. 176, 5378-5384.

Marcus, M., and Halpern, Y.S. (1969). Genetic and physiologic analysis of glutamate decarboxylase in Escherichia coli K-12. J.Bacteriol. 97, 1509-1510.

Marschall, C., and Hengge-Aronis, R. (1995). Regulatory characteristics and promoter analysis of csiE, a stationary phase-inducible gene under the control of sigma S and the cAMP-CRP complex in Escherichia coli. Molec.Microbiol. 18, 175-184.

Marschall, C., Labrousse, V., Kreimer, M., Weichart, D., Ko1b, A, and Hengge-Aronis, R. (1998). Molecular analysis ofthe regulation of csiD, a carbon starvation­ inducible gene in Escherichia coli that is exclusively dependent on sigma S and requires activation by cAMP-CRP. J.MoLBioL 276, 339-353.

Metzer, E., and Halpern, Y.S. (1990). In vivo cloning and characterization ofthe gabCTDP gene cluster ofEscherichia coli K-12. J.BacterioL 172, 3250-3256.

Metzer, E., Levitz, R., and Halpern, Y.S. (1979). Isolation and properties ofEscherichia coli K-12 mutants impaired in the utilization of gamma-amino butyrate. J.Bacteriol. 137,1111-1118.

Miksch, G. and Dobrowolski, P. (1995). Growth phase-dependent induction of stationary-phase promoters ofEscherichia coli in different gram-negative bacteria. J.Bacteriol. 177, 5374-5378.

Miller, J.H. (1992). A short course in bacterial genetics. A laboratory manual and handbook for Escherichia coli and related bacteria. Cold Spring Harbor Press, Cold Spring Harbor, New York. 92

Mulvey, M.R and Loewen, P.C. (1989). Nucleotide sequence of katF of Escherichia coli suggests KatF protein is a novel sigma transcription factor. Nucleic Acids Res. 17, 9979-9991.

Neidhardt, F.C., and Magasanik, B. (1957). Reversal of the glucose inhibition of histidase synthesis in Aerobacter aerogenes. J.Bacteriol. 73, 253-259.

Niegemann, E., Schulz, A, and Bartsch, K. (1993). Molecular organization of the Escherichia coli gab cluster: nucleotide sequence ofthe structural genes gabD and gabP and expression ofthe GABA permease gene. Arch.Microbiol. 160, 454-460.

Nirenberg, M.W., and Jakoby, W.B. (1960). Enzymatic utilization ofy-hydroxybutyric acid. J.Biol.Chem. 235, 954-960.

Paiva, E.T., and Silhavy, T.J. (1984). LacZfusions to genes that specify exported proteins: a general technique. Mol. Gen. Genet. 194, 388-394.

Pickett, M., Gwynne, D.l., Buxton, F.P., Elliott, R, Davies, RW., Lockington, RA, Scazzocchio, C., Sealy-Lewis, H.M. (1987). Cloning and characterization of the aidA gene ofAspergillus nidulans. Gene 51,217-226.

Reeve, C.A., Amy, P.S., and Matin, A (1984). Role of protein synthesis in the survival of carbon-starved Escherichia coli K-12. J.Bacteriol. 160, 1041-1046.

Roberts, E., and Frankel, S. (1950). y-Aminobutyric acid in brain: its formation from glutamic acid. J.Biol.Chem. 187, 55-63.

Schellhorn, H.E., Audia, J.P., Wei, L.I.C., and Chang, L. (1998). Identification of conserved, RpoS-dependent stationary-phase genes of Escherichia coli. J.Bacteriol 180, 6283-6291.

Schulz, A, Taggeselle, P., Tripier, D., and Bartsch, K. (1990). Stereospecific production ofthe herbicide phosphinothricin (glufosinate) by transamination: isolation and characterization of a phosphinothricin-specific transaminase from Escherichia coli. Appl.Environ.Microbiol. 56, 1-6.

Scott, E.M., and Jakoby, W.B. (1959). Soluble y-aminobutyric-glutamic transaminase from Pseudomonas fluorescens. J.Biol.Chem. 234, 932-936.

Shat"be, E., Metzer, E., and Halpern, Y.S. (1985). Metabolic pathway for the utilization 93

ofL-arginine, L-ornithine, agmatine and putrescine as nitrogen sources in Escherichia coli K-12. J.BacterioL 163, 933-931.

Small, P., Blankenhorn, D., Welty, D., Zinser, E., and Slonczewski, J.L. (1994). Acid and base resistance in Escherichia coli and Shigella jlexneri: role of rpoS and growthpH. J.BacterioL 176, 1729-1737.

Smi~ G.R, and Magasanik, B. (1971b). The two operons of the histidine utilization system in Salmonella typhimurium. J.BioLChem. 246,3330-3341.

Smi~ G.R, Halpern, Y.S., and Magasanik, B. (1971a). Genetic and metabolic control of enzymes responsible for histidine degradation in Salmonella typhimurium: 4- imidazolone-5-propionate amidohydrolase and N-fonnimino-L-glutamate formiminohydrolase. J.BioLChem. 246, 3320-3329.

Sophianopoulou, V., and Scazzocchio, C. (1989). The proline transport protein of Aspergillus nidulans is very similar to amino acid transporters ofSaccharomyces cerevisiae. Mol.Microbiol. 3, 105-114.

Stern, M.J., Higgins, C.F., and Ames, G.F.-L. (1984). Isolation and characterization of lac fusions to two nitrogen-regulated promoters. Mol.Gen.Genet. 195, 219-227.

Tanaka, J., and Fink, G.R (1985). The histidine permease gene (HIP1) of Saccharomyces cerevisiae. Gene 38,205-214.

Tyler, B. (1978). Regulation ofthe assimilation of nitrogen compounds. Annu. Rev. Biochem. 47, 1127-1162.

Tyler, B., DeLeo, AB., and Magasanik, B. (1974). Activation of transcription of hut DNA by glutamine synthetase. Proc.NatlAcad.Sci U.S.A 71,225-229.

Vandenbol, M., Jauniaux, J.C., and Grenson, M. (1989). Nucleotide sequence ofthe Saccharomyces cerevisiae PUT4 proline-permease-encoding gene: similarities between CANl, HIP1 and PUT4 permeases. Gene 83, 153-159.

Wallace, B., Yang, Y.J., Hong, J.S., and Lurn, D. (1990). Cloning and sequencing of a gene encoding a glutamate and aspartate carrier ofEscherichia coli K-12. J.Bacteriol. 172, 3214-3220.

Willis, RC., Iwata, K.K., and Furlong, C.E. (1975). Regulation of glutamine transport in Escherichia coli. J.Bacteriol. 122, 1032-1037. 94

Xu, J. and Johnson, RC. (1995). Identification of genes negatively regulated by Fis: Fis and RpoS comodulate growth-phase-dependent gene expression in Escherichia coli. J.Bacteriol 177, 938-947.

Yonaha, K., and Yoyama, S. (1980). Gamma-aminobutyrate:alpha-ketoglutarate aminotransferase :from Pseudomonas sp. F-126: purification, crystallization, and enzymologic properties. Arch.Biochem.Biophys. 200, 156-164.

Yonaha, K., Suzuki, K., and Toyama, S. (1985). 4-aminobutyrate: 2-oxoglutarate aminotransferase ofStreptomyces griseus: purification and properties. Eur.J.Biochem. 146, 101-106.

Zaboura, M., and Halpern, Y.S. (1978). Regulation ofy-aminobutyric acid degradation in Escherichia coli by nitrogen metabolism enzymes. J.Bacteriol 133, 447-451.